INDEX

Explanations

specific mentions of the object or concept 'a'

oai_token-act-pair · gpt-3.5-turbo

the article "a" and other related indefinite articles

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

etsk

-0.85

 encour

-0.82

ATURES

-0.75

 outputs

-0.72

 credits

-0.71

words

-0.70

reports

-0.70

Examples

-0.70

 entries

-0.70

iments

-0.70

POSITIVE LOGITS

 bunch

0.92

 dozen

0.88

 gigantic

0.88

 particular

0.86

rouse

0.84

 nutshell

0.84

 handful

0.84

 small

0.82

 large

0.82

 certain

0.80

Activations Density 0.226%

specific mentions of the object or concept 'a'

the article "a" and other related indefinite articles

No Comments

No Known Activations