INDEX

Explanations

information about documents, articles, or publications that include specific elements or details

oai_token-act-pair · gpt-3.5-turbo

key terms related to formal documents or proposals

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

assetsadobe

-0.90

esters

-0.79

Games

-0.76

luck

-0.75

Shop

-0.71

projects

-0.70

Mania

-0.69

chool

-0.68

Stud

-0.68

ateurs

-0.68

POSITIVE LOGITS

 mention

1.43

 disclaimer

1.41

 references

1.39

 caveat

1.27

 clause

1.26

 reference

1.20

 explicit

1.19

 quotation

1.15

 clauses

1.14

 mentions

1.08

Activations Density 0.304%

information about documents, articles, or publications that include specific elements or details

key terms related to formal documents or proposals

No Comments

No Known Activations