INDEX

Explanations

phrases indicating high probability or likelihood of an event happening

oai_token-act-pair · gpt-3.5-turbo

phrases indicating probability or likelihood

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

gado

-0.73

zeb

-0.71

elta

-0.71

inth

-0.71

ente

-0.70

enced

-0.70

hips

-0.70

ells

-0.67

gencies

-0.66

Ã§a

-0.66

POSITIVE LOGITS

 underestimate

0.83

 releg

0.78

 assume

0.76

 rely

0.74

 equate

0.71

 partName

0.70

 infer

0.70

entimes

0.69

 unfocusedRange

0.69

 underestimated

0.68

Activations Density 0.024%

phrases indicating high probability or likelihood of an event happening

phrases indicating probability or likelihood

No Comments

No Known Activations