INDEX

Explanations

references to skiing and resort-related features

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

/tos

-0.17

nge

-0.16

_kw

-0.15

rÃ¡t

-0.15

Ã´n

-0.14

Äł

-0.14

ileÅŁ

-0.14

vens

-0.13

 Ð±Ð»

-0.13

etsy

-0.13

POSITIVE LOGITS

ski

0.53

 skiing

0.48

Ski

0.48

 snow

0.42

ski

0.39

 powder

0.38

SKI

0.37

 Powder

0.36

sk

0.36

 Snow

0.35

Activations Density 0.053%

references to skiing and resort-related features

No Comments

No Known Activations