INDEX

Explanations

mentions of specific episodes of TV shows or podcasts

oai_token-act-pair · gpt-3.5-turbo

references to episodes of television shows or podcasts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

cffffcc

-0.76

 edges

-0.73

headed

-0.71

anders

-0.70

istas

-0.68

 entrances

-0.68

ettlement

-0.67

ices

-0.66

 engulf

-0.66

chwitz

-0.66

POSITIVE LOGITS

 Weekly

0.83

 Forbes

0.82

ABC

0.79

 Monthly

0.79

 Premiere

0.79

 Popular

0.75

PBS

0.74

 SHOW

0.74

 Trend

0.74

 Keeping

0.73

Activations Density 0.121%

mentions of specific episodes of TV shows or podcasts

references to episodes of television shows or podcasts

No Comments

No Known Activations