INDEX

Explanations

phrases related to grammar and proper writing

oai_token-act-pair · gpt-3.5-turbo

references to grammar-related concepts or issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

kus

-0.76

;;

-0.72

zee

-0.70

ership

-0.70

ply

-0.68

;;;;;;;;

-0.68

ming

-0.68

izen

-0.67

immer

-0.66

cession

-0.66

POSITIVE LOGITS

otle

0.96

uracy

0.80

 Nieto

0.70

dayName

0.65

apor

0.64

 Camb

0.64

grad

0.63

ritic

0.62

ravel

0.61

 Aval

0.61

Activations Density 0.031%

phrases related to grammar and proper writing

references to grammar-related concepts or issues

No Comments

No Known Activations