INDEX

Explanations

phrases related to movement or travel

oai_token-act-pair · gpt-3.5-turbo

phrases indicating origin or source

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

attribute

-0.84

natureconservancy

-0.79

achy

-0.74

faced

-0.73

sylv

-0.73

few

-0.73

wolves

-0.71

eous

-0.70

ema

-0.70

heet

-0.70

POSITIVE LOGITS

 afar

0.96

 abroad

0.91

 whence

0.88

 thence

0.85

 Wichita

0.80

0.77

 Spokane

0.77

 Guatemala

0.76

 Palest

0.75

 Zurich

0.74

Activations Density 0.153%

phrases related to movement or travel

phrases indicating origin or source

No Comments

No Known Activations