INDEX

Explanations

nouns and verbs that relate to actions and characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

itar

-0.15

æķ·

-0.14

å®ħ

-0.14

toolbox

-0.14

edy

-0.13

ussed

-0.13

anic

-0.13

ischen

-0.13

 Unary

-0.13

oho

-0.13

POSITIVE LOGITS

bote

0.18

~/

0.16

readcr

0.14

DMI

0.14

ãĥ¼ãĥł

0.13

Ã¸

0.13

_owned

0.13

ovÃ¡no

0.13

rame

0.13

_Tis

0.13

Activations Density 0.072%

nouns and verbs that relate to actions and characteristics

No Comments

No Known Activations