INDEX

Explanations

adjectives and their usage in descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

rint

-0.17

uhan

-0.17

oeff

-0.16

Inlining

-0.14

umbnails

-0.14

arih

-0.14

izzer

-0.14

ugu

-0.14

ixels

-0.14

EMPL

-0.14

POSITIVE LOGITS

áŀ¶

0.17

edImage

0.15

Gr

0.14

 Morrow

0.14

sel

0.14

 Stim

0.14

0.13

/accounts

0.13

ãĥ¼ãĥģ

0.13

Activations Density 0.047%

adjectives and their usage in descriptions

No Comments

No Known Activations