INDEX

Explanations

words related to testing or evaluation

oai_token-act-pair · gpt-3.5-turbo

words related to testing and evaluation

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

 parcels

-0.67

Ô

-0.67

eret

-0.65

 tracts

-0.64

lance

-0.64

 Levant

-0.63

Scotland

-0.63

ĸļ

-0.62

aucas

-0.62

 Territories

-0.60

POSITIVE LOGITS

 hypotheses

0.84

prep

0.78

formance

0.76

MRI

0.71

 readiness

0.70

grad

0.69

 viability

0.68

ickson

0.66

 hypothesis

0.66

tle

0.66

Activations Density 0.248%

words related to testing or evaluation

words related to testing and evaluation

No Comments

No Known Activations