INDEX

Explanations

phrases related to starting events or actions

oai_token-act-pair · gpt-3.5-turbo

phrases related to starting points and progress 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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

yg

-0.71

ACTED

-0.66

hai

-0.63

SPONSORED

-0.60

xual

-0.60

enance

-0.59

luent

-0.58

CVE

-0.58

wolves

-0.58

clud

-0.57

POSITIVE LOGITS

 rocky

0.79

oots

0.71

lopp

0.68

urous

0.67

 merry

0.66

 tails

0.63

 rave

0.62

 Vaughan

0.62

 adventures

0.60

rocal

0.60

Activations Density 0.083%

phrases related to starting events or actions

phrases related to starting points and progress in various contexts

No Comments

No Known Activations