INDEX

Explanations

words related to proficiency or skill in a specific task or activity

oai_token-act-pair · gpt-3.5-turbo

phrases indicating skill or competence

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

ships

-0.69

shell

-0.68

loo

-0.65

GROUP

-0.65

fill

-0.64

arch

-0.63

trip

-0.62

bender

-0.62

cott

-0.61

plate

-0.61

POSITIVE LOGITS

 least

0.98

abase

0.88

roph

0.86

onement

0.84

 conception

0.80

omics

0.80

 predicting

0.77

 detecting

0.76

 estimating

0.75

 recognizing

0.75

Activations Density 0.093%

words related to proficiency or skill in a specific task or activity

phrases indicating skill or competence

No Comments

No Known Activations