INDEX

Explanations

phrases related to mental processes and thoughts

oai_token-act-pair · gpt-3.5-turbo

words related to thoughts and mental processes

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

 Firm

-0.67

 Moder

-0.66

 Pist

-0.63

afort

-0.62

byn

-0.62

 degrees

-0.59

ammy

-0.59

Pix

-0.56

 Norm

-0.56

hi

-0.55

POSITIVE LOGITS

steps

0.85

hole

0.79

swing

0.67

 selves

0.67

INESS

0.65

 cavity

0.65

eks

0.63

 doorstep

0.63

PATH

0.62

 balcony

0.60

Activations Density 0.140%

phrases related to mental processes and thoughts

words related to thoughts and mental processes

No Comments

No Known Activations