INDEX

Explanations

concepts related to legal proceedings and liability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

agrams

-0.18

ÑĮÐºÐ¾

-0.16

Ð¶Ð¸

-0.16

ronym

-0.15

/GPL

-0.15

arent

-0.14

agram

-0.14

è´¨

-0.14

utterstock

-0.14

è³ª

-0.14

POSITIVE LOGITS

 damages

0.45

 awards

0.41

 award

0.39

dam

0.39

 recovery

0.39

 awarded

0.38

 recover

0.38

Dam

0.36

 Awards

0.35

 damage

0.35

Activations Density 0.327%

concepts related to legal proceedings and liability

No Comments

No Known Activations