INDEX

Explanations

references to specific individuals and organizations

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

ilo

-0.15

 Norm

-0.15

ules

-0.14

izzo

-0.14

ex

-0.14

ensus

-0.14

conda

-0.13

ph

-0.13

 Cous

-0.13

mer

-0.13

POSITIVE LOGITS

0.15

ãģ¡ãģ¯

0.15

lal

0.14

ģµ

0.14

uzzle

0.14

aklÄ±

0.14

etto

0.14

UIL

0.14

Ĥ¬

0.13

 witness

0.13

Activations Density 0.008%

references to specific individuals and organizations

No Comments

No Known Activations