INDEX

Explanations

terms related to capabilities and abilities

oai_token-act-pair · gpt-4o-mini Triggered by @bot

the word "ability" or "Ability" (especially when followed by "to").

oai_token-act-pair · claude-4-5-sonnet Triggered by @jyhe0408

expressions that denote capability or capacity, especially in “ability to …” constructions.

oai_token-act-pair · gpt-5 Triggered by @jyhe0408

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ilion

-0.17

ÐµÐ¼Ð°

-0.16

olta

-0.15

mers

-0.15

izoph

-0.15

ers

-0.15

ê

-0.15

celik

-0.15

ilyn

-0.15

role

-0.14

POSITIVE LOGITS

-bodied

0.31

/disable

0.19

andex

0.16

idades

0.16

hood

0.16

uali

0.16

keiten

0.16

 Hurricane

0.16

åĲ¦

0.16

ustin

0.15

Activations Density 0.056%

terms related to capabilities and abilities

the word "ability" or "Ability" (especially when followed by "to").

expressions that denote capability or capacity, especially in “ability to …” constructions.

No Comments

No Known Activations