INDEX

Explanations

phrases or words indicating correctness or accuracy

oai_token-act-pair · gpt-3.5-turbo

references to accuracy and correctness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

CHO

-0.74

aden

-0.72

atos

-0.72

 Valhalla

-0.70

GGGGGGGG

-0.70

fleet

-0.67

EMOTE

-0.67

Das

-0.63

belt

-0.63

atten

-0.63

POSITIVE LOGITS

ives

0.90

eous

0.85

yt

0.84

able

0.84

ibly

0.84

ible

0.81

 spelling

0.80

 guiIcon

0.79

itude

0.78

 answers

0.77

Activations Density 0.016%

phrases or words indicating correctness or accuracy

references to accuracy and correctness

No Comments

No Known Activations