INDEX

Explanations

proper nouns, specifically names and brand references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

UseProgram

-0.15

TEGER

-0.14

quals

-0.14

venile

-0.14

ecies

-0.14

ÑĢÐ¸Ð·

-0.14

DebugEnabled

-0.14

á»ĳ

-0.14

olian

-0.14

rosso

-0.13

POSITIVE LOGITS

 Bret

0.35

 Brett

0.34

 Gret

0.34

CRT

0.31

CRT

0.31

Rut

0.30

 Witt

0.30

(Qt

0.30

 Holt

0.30

agt

0.29

Activations Density 0.735%

proper nouns, specifically names and brand references

No Comments

No Known Activations