INDEX

Explanations

phrases indicating actions that can be performed or completed

oai_token-act-pair · gpt-3.5-turbo

the verb "be" in various forms, indicating existence or state of being

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

 Kamp

-0.69

Rab

-0.69

-0.68

rones

-0.65

 Borders

-0.65

 Blitz

-0.63

 Notting

-0.63

Rus

-0.61

Buc

-0.59

 Bans

-0.59

POSITIVE LOGITS

acons

0.95

 likened

0.92

 traced

0.90

 forgiven

0.89

 seen

0.88

leeve

0.88

 construed

0.87

 viewed

0.86

 undone

0.83

 easily

0.83

Activations Density 0.111%

phrases indicating actions that can be performed or completed

the verb "be" in various forms, indicating existence or state of being

No Comments

No Known Activations