INDEX

Explanations

specific conditions or qualifiers related to entities within a sentence

oai_token-act-pair · gpt-3.5-turbo

conditional phrases or clauses indicating specific types of stories or situations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Cheong

-0.78

ibaba

-0.70

 Rabbit

-0.66

thing

-0.65

åĤ

-0.63

 Belt

-0.63

 Lauder

-0.62

bay

-0.62

.............

-0.61

Balt

-0.61

POSITIVE LOGITS

 violate

0.88

 specialize

0.87

 exceed

0.85

 stray

0.82

 contain

0.81

 explicitly

0.78

 originate

0.75

affected

0.75

 utilize

0.74

 emit

0.74

Activations Density 0.285%

specific conditions or qualifiers related to entities within a sentence

conditional phrases or clauses indicating specific types of stories or situations

No Comments

No Known Activations