INDEX

Explanations

references to specific movie studios

oai_token-act-pair · gpt-3.5-turbo

mentions of film studios and related entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

theless

-0.74

mere

-0.69

ucl

-0.68

bh

-0.67

pend

-0.65

stal

-0.65

 Supporters

-0.64

isms

-0.63

osate

-0.63

abad

-0.63

POSITIVE LOGITS

 studio

0.89

 studios

0.87

yrinth

0.80

mates

0.79

 albums

0.77

 apartment

0.73

 floor

0.73

orpor

0.73

floor

0.72

 mates

0.70

Activations Density 0.036%

references to specific movie studios

mentions of film studios and related entities

No Comments

No Known Activations