INDEX

Explanations

words related to scientific or medical testing procedures

oai_token-act-pair · gpt-3.5-turbo

references to evaluations or assessments, specifically tests

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ilan

-0.78

rir

-0.69

esan

-0.63

ominated

-0.62

DonaldTrump

-0.61

iture

-0.60

rue

-0.59

tsy

-0.59

chant

-0.59

 Tradable

-0.58

POSITIVE LOGITS

 tests

3.86

 Tests

3.22

tests

2.91

 test

2.64

 testing

2.57

 Testing

2.25

Testing

2.17

 tested

2.10

testing

2.09

 testers

2.08

Activations Density 0.016%

words related to scientific or medical testing procedures

references to evaluations or assessments, specifically tests

No Comments

No Known Activations