INDEX

Explanations

references to walking and outdoor activities

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

ajs

-0.16

eel

-0.15

ÑĢÑĮ

-0.15

ldb

-0.15

UNCTION

-0.14

unker

-0.14

ebe

-0.14

 boarding

-0.13

unction

-0.13

ween

-0.13

POSITIVE LOGITS

 walks

0.54

 walk

0.49

walk

0.45

 Walk

0.43

Walk

0.42

 walked

0.41

_walk

0.40

 walking

0.39

.walk

0.36

 walker

0.35

Activations Density 0.116%

references to walking and outdoor activities

No Comments

No Known Activations