INDEX

Explanations

words related to proteins

oai_token-act-pair · gpt-3.5-turbo

references to protein in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Fantasy

-0.71

rolet

-0.70

orld

-0.69

 Noon

-0.69

 FANTASY

-0.64

sbm

-0.63

ewitness

-0.63

quished

-0.63

 Faust

-0.62

arters

-0.62

POSITIVE LOGITS

aceous

1.17

rotein

0.98

zyme

0.97

protein

0.96

 protein

0.94

 isolate

0.89

 powd

0.89

 Protein

0.87

 proteins

0.87

uria

0.86

Activations Density 0.010%

words related to proteins

references to protein in various contexts

No Comments

No Known Activations