INDEX

Explanations

phrases related to decision-making or evaluation

oai_token-act-pair · gpt-3.5-turbo

expressions of obligation or expectation

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

>>>>>>>>

-0.59

reports

-0.49

Zup

-0.48

theless

-0.47

bilt

-0.46

////////////////

-0.45

vertisement

-0.45

eers

-0.44

++++++++

-0.44

udos

-0.42

POSITIVE LOGITS

 otherwise

0.49

 supposedly

0.44

 someday

0.42

 traditionally

0.42

 racially

0.42

 vulner

0.41

akeru

0.41

 normally

0.41

 previously

0.41

 perceive

0.40

Activations Density 0.925%

phrases related to decision-making or evaluation

expressions of obligation or expectation

No Comments

No Known Activations