INDEX

Explanations

articles ('a', 'an', 'the') followed by single nouns

oai_token-act-pair · gpt-3.5-turbo

instances of the indefinite article "a" or "an" and their contextual associations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Sources

-0.76

 Jagu

-0.74

Links

-0.73

inson

-0.70

odore

-0.69

AUD

-0.65

aci

-0.65

 aliases

-0.64

killers

-0.64

 Amend

-0.64

POSITIVE LOGITS

 particular

1.17

 bunch

1.09

 person

1.08

 certain

1.02

lot

1.01

uras

0.99

rouse

0.96

 stranger

0.92

 single

0.90

 piece

0.90

Activations Density 0.306%

articles ('a', 'an', 'the') followed by single nouns

instances of the indefinite article "a" or "an" and their contextual associations

No Comments

No Known Activations