INDEX

Explanations

mentions of the size or scale of objects or events

oai_token-act-pair · gpt-3.5-turbo

significant career changes and turning points in a narrative

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

emort

-0.77

Hes

-0.70

thereum

-0.69

arov

-0.69

neum

-0.68

rue

-0.65

 Flore

-0.63

swick

-0.62

aunder

-0.61

arians

-0.61

POSITIVE LOGITS

Big

1.93

big

1.90

big

1.88

 Small

1.86

 smaller

1.86

small

1.86

Big

1.85

Small

1.84

BIG

1.84

 small

1.81

Activations Density 0.357%

mentions of the size or scale of objects or events

significant career changes and turning points in a narrative

No Comments

No Known Activations