INDEX

Explanations

locations and starting or ending points

oai_token-act-pair · gpt-3.5-turbo

phrases related to beginnings, transitions, and expansions 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Gw

-0.68

usercontent

-0.63

 yuan

-0.63

Tam

-0.61

eland

-0.60

udos

-0.59

Mug

-0.59

ubs

-0.58

efe

-0.57

ilty

-0.57

POSITIVE LOGITS

 secondly

1.23

 followed

1.10

 thereafter

1.06

Secondly

0.97

 foremost

0.97

 gradually

0.95

 culminating

0.92

 Secondly

0.88

 progressed

0.87

 then

0.86

Activations Density 0.542%

locations and starting or ending points

phrases related to beginnings, transitions, and expansions in various contexts

No Comments

No Known Activations