INDEX

Explanations

text and symbols related to a specific formatting or coding convention

oai_token-act-pair · gpt-3.5-turbo

nouns and technical terminology related to complex systems and probabilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

•Partition Features Layer 8 GPT2 Small

" Token" vs "token"

No Comments

Negative Logits

 Haku

-0.81

 Tent

-0.77

Mub

-0.76

 Liter

-0.74

 Robot

-0.74

Yad

-0.73

 Amon

-0.71

 Tribal

-0.70

 Tier

-0.70

Sob

-0.70

POSITIVE LOGITS

together

1.23

ó

1.22

Loading

1.21

without

1.20

administ

1.17

matter

1.17

within

1.14

subject

1.14

Commission

1.13

enough

1.13

Activations Density 0.202%

text and symbols related to a specific formatting or coding convention

nouns and technical terminology related to complex systems and probabilities

No Comments

No Known Activations