INDEX

Explanations

references to trial-related terminology and legal procedures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

eland

-0.21

eding

-0.17

ements

-0.17

auce

-0.16

ema

-0.16

iel

-0.16

es

-0.15

aos

-0.15

_squared

-0.15

onomic

-0.15

POSITIVE LOGITS

ogue

0.29

 nghiá»ĩm

0.26

ounge

0.17

orca

0.16

onga

0.16

/test

0.16

/Test

0.16

ists

0.15

tech

0.15

ld

0.15

Activations Density 0.014%

references to trial-related terminology and legal procedures

No Comments

No Known Activations