INDEX

Explanations

the word "refrain"

oai_token-act-pair · gpt-3.5-turbo

the word "refrain" used in various contexts, particularly related to avoiding certain actions or behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

NetMessage

-0.84

rations

-0.78

ammy

-0.76

rams

-0.70

immer

-0.69

ramid

-0.69

neau

-0.67

 Pike

-0.67

odes

-0.63

opic

-0.63

POSITIVE LOGITS

 refrain

1.39

 abst

0.84

rences

0.84

hovah

0.76

 acknow

0.75

fulness

0.74

SourceFile

0.73

 attRot

0.72

 restraint

0.68

 compe

0.66

Activations Density 0.011%

the word "refrain"

the word "refrain" used in various contexts, particularly related to avoiding certain actions or behaviors

No Comments

No Known Activations