INDEX

Explanations

scenarios or hypothetical situations

oai_token-act-pair · gpt-3.5-turbo

descriptive phrases about various hypothetical scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ighters

-0.94

hammad

-0.87

ove

-0.84

anguages

-0.84

emouth

-0.83

ixed

-0.82

olulu

-0.81

alties

-0.81

oves

-0.80

itsch

-0.80

POSITIVE LOGITS

 scenario

1.03

 scenarios

1.01

 involving

0.80

 unfold

0.78

 probabilities

0.70

 Heller

0.70

0.69

 unfolding

0.68

eers

0.67

0.65

Activations Density 0.015%

scenarios or hypothetical situations

descriptive phrases about various hypothetical scenarios

No Comments

No Known Activations