INDEX

Explanations

locations or places mentioned in the text

oai_token-act-pair · gpt-3.5-turbo

the word "in" across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Upload

-0.64

giving

-0.60

 critics

-0.59

nell

-0.59

-0.57

enaries

-0.55

linger

-0.54

 chops

-0.54

 detractors

-0.54

¿½

-0.54

POSITIVE LOGITS

 sight

0.98

 existence

0.94

between

0.93

 society

0.87

animate

0.83

planet

0.81

 life

0.81

 heaven

0.79

 between

0.78

 history

0.77

Activations Density 0.116%

locations or places mentioned in the text

the word "in" across various contexts

No Comments

No Known Activations