INDEX

Explanations

appearances of film-related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "film."

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

bley

-0.83

pron

-0.71

 overloaded

-0.70

lies

-0.70

ppard

-0.69

attled

-0.67

ensional

-0.66

 tooth

-0.66

umbn

-0.65

 punishing

-0.63

POSITIVE LOGITS

 Film

1.17

 Films

1.15

Film

1.13

film

1.04

 Noir

0.97

 Festival

0.90

 Cinema

0.87

Lens

0.87

 Productions

0.87

 Animation

0.84

Activations Density 0.012%

appearances of film-related terms

mentions of the word "film."

No Comments

No Known Activations