INDEX

Explanations

statements related to disclosure or justification

oai_token-act-pair · gpt-3.5-turbo

discussions centered around honesty and revelation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

phalt

-0.76

rians

-0.70

atl

-0.70

croft

-0.68

agues

-0.67

rian

-0.66

erva

-0.66

iatrics

-0.65

enf

-0.65

onder

-0.64

POSITIVE LOGITS

his

1.07

 their

1.03

 herself

0.93

her

0.89

 owning

0.87

 what

0.86

how

0.86

their

0.86

 himself

0.85

 quitting

0.85

Activations Density 0.401%

statements related to disclosure or justification

discussions centered around honesty and revelation

No Comments

No Known Activations