INDEX

Explanations

words related to shape or transformation

oai_token-act-pair · gpt-3.5-turbo

references to various shapes and forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Mub

-0.79

govtrack

-0.76

onte

-0.74

iance

-0.71

 Edited

-0.70

amily

-0.67

uers

-0.65

endment

-0.65

ondo

-0.64

GS

-0.64

POSITIVE LOGITS

 shif

0.98

shape

0.93

 shape

0.87

 shapes

0.78

ly

0.76

fitting

0.75

cut

0.74

 Shape

0.72

sheet

0.72

lier

0.72

Activations Density 0.027%

words related to shape or transformation

references to various shapes and forms

No Comments

No Known Activations