INDEX

Explanations

references to unspecified or mysterious entities or situations

oai_token-act-pair · gpt-3.5-turbo

references to unidentified or unspecified entities and events

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

boa

-0.89

ickr

-0.81

igsaw

-0.78

odcast

-0.78

utic

-0.75

audi

-0.75

ongyang

-0.74

ŃĶ

-0.74

apers

-0.74

ixels

-0.73

POSITIVE LOGITS

 quantity

0.83

theless

0.80

 unknown

0.76

Origin

0.76

 Mortal

0.73

terday

0.72

comings

0.72

erness

0.70

 jurisdiction

0.65

landish

0.64

Activations Density 0.025%

references to unspecified or mysterious entities or situations

references to unidentified or unspecified entities and events

No Comments

No Known Activations