INDEX

Explanations

phrases related to personal preferences and tastes

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

orges

-0.16

umba

-0.15

Ã¼zel

-0.14

 ActionTypes

-0.14

à¤Ľ

-0.14

 labore

-0.14

ripp

-0.14

obil

-0.13

rous

-0.13

opyright

-0.13

POSITIVE LOGITS

 taste

0.74

 tastes

0.69

 Taste

0.56

 tasted

0.44

 preferences

0.42

 tast

0.39

 Ð²ÐºÑĥ

0.39

 preference

0.37

 liking

0.33

preferences

0.33

Activations Density 0.085%

phrases related to personal preferences and tastes

No Comments

No Known Activations