INDEX

Explanations

phrases or clauses indicating a past action or state

oai_token-act-pair · gpt-3.5-turbo

phrases indicating past habits or repeated actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

essional

-0.75

edIn

-0.69

sequence

-0.67

 Compass

-0.67

Choice

-0.66

stakes

-0.64

lations

-0.63

metics

-0.63

ulus

-0.62

starter

-0.62

POSITIVE LOGITS

 shrink

0.77

 stomp

0.75

 haunt

0.73

 numbering

0.72

 advertise

0.72

 crush

0.71

 scoop

0.70

 treat

0.70

iaries

0.68

 mill

0.68

Activations Density 0.034%

phrases or clauses indicating a past action or state

phrases indicating past habits or repeated actions

No Comments

No Known Activations