INDEX

Explanations

references to geographical locations or proper nouns

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

ombre

-0.19

arr

-0.17

ames

-0.17

ikh

-0.16

ear

-0.15

PR

-0.15

ude

-0.15

AME

-0.14

icked

-0.14

amed

-0.14

POSITIVE LOGITS

gor

0.19

antes

0.18

ETHER

0.17

anking

0.17

ort

0.16

Ã¼rnberg

0.16

jar

0.16

ettle

0.16

Ð¸Ð¶

0.16

ords

0.16

Activations Density 0.033%

references to geographical locations or proper nouns

No Comments

No Known Activations