INDEX

Explanations

instances of people taking actions or expressing opinions

oai_token-act-pair · gpt-3.5-turbo

numerical data or statistical references

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

 Advertisement

-0.63

anwhile

-0.57

berus

-0.50

externalActionCode

-0.50

 Continued

-0.50

direction

-0.49

 Lavrov

-0.48

low

-0.47

_-

-0.47

 Recap

-0.47

POSITIVE LOGITS

onite

0.70

 theirs

0.63

 themselves

0.57

 collectively

0.57

icable

0.56

 their

0.54

 converge

0.53

chwitz

0.53

 together

0.52

amily

0.51

Activations Density 1.036%

instances of people taking actions or expressing opinions

numerical data or statistical references

No Comments

No Known Activations