INDEX

Explanations

mentions of dressing rooms

oai_token-act-pair · gpt-3.5-turbo

references to dressing or related activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

founded

-0.74

ntil

-0.72

ellen

-0.72

fram

-0.71

reme

-0.71

untu

-0.71

sighted

-0.70

ecd

-0.70

ItemTracker

-0.70

JV

-0.67

POSITIVE LOGITS

 dressing

0.93

 gown

0.92

 rooms

0.92

glers

0.91

 rehearsal

0.83

ptin

0.82

rooms

0.76

 rehears

0.75

room

0.75

 Diesel

0.72

Activations Density 0.003%

mentions of dressing rooms

references to dressing or related activities

No Comments

No Known Activations