INDEX

Explanations

phrases indicating location or proximity to a specific place

oai_token-act-pair · gpt-3.5-turbo

the conjunctions "and" and "or" as they relate to various contexts

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

 Welch

-0.62

 Tasman

-0.62

 Examiner

-0.61

 Hung

-0.60

FAT

-0.59

 Holl

-0.57

 Pric

-0.56

Fat

-0.56

 Sham

-0.55

 Cain

-0.55

POSITIVE LOGITS

 around

1.14

around

1.02

rogen

0.99

near

0.99

rogens

0.96

 near

0.89

 outside

0.84

within

0.81

fro

0.81

 surrounding

0.80

Activations Density 0.051%

phrases indicating location or proximity to a specific place

the conjunctions "and" and "or" as they relate to various contexts

No Comments

No Known Activations