INDEX

Explanations

words related to concealment or secrecy

oai_token-act-pair · gpt-3.5-turbo

instances of the word "hide" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ombat

-0.80

ctive

-0.78

ammy

-0.76

onian

-0.69

ersive

-0.69

oker

-0.69

orough

-0.68

union

-0.67

signed

-0.67

FK

-0.66

POSITIVE LOGITS

ously

1.03

away

0.82

hid

0.80

 hide

0.79

 hides

0.75

 behind

0.75

 hiding

0.75

Clo

0.75

rets

0.73

 secrets

0.71

Activations Density 0.026%

words related to concealment or secrecy

instances of the word "hide" and its variations

No Comments

No Known Activations