INDEX

Explanations

references to secretive or stealthy behavior such as "sneak," "sneaky," or "crept."

oai_token-act-pair · gpt-3.5-turbo

words related to stealth or sneaky actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Pwr

-0.80

relations

-0.74

nan

-0.72

onial

-0.71

upon

-0.69

ricular

-0.68

rices

-0.68

rica

-0.67

ternity

-0.67

verend

-0.67

POSITIVE LOGITS

 sneak

1.16

ily

1.01

 peek

0.94

 glances

0.89

door

0.81

ipeg

0.77

atures

0.77

 sneaking

0.76

ously

0.72

 Sneak

0.69

Activations Density 0.008%

references to secretive or stealthy behavior such as "sneak," "sneaky," or "crept."

words related to stealth or sneaky actions

No Comments

No Known Activations