INDEX

Explanations

phrases related to changes or transitions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a change or transition in status or condition

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

iosyn

-0.81

QUI

-0.78

jong

-0.77

cise

-0.76

resy

-0.70

vic

-0.67

DAQ

-0.66

ourage

-0.66

etimes

-0.64

Enlarge

-0.64

POSITIVE LOGITS

 obscurity

0.90

 hating

0.80

 afar

0.73

 humble

0.71

 mildly

0.66

 whence

0.66

 dormant

0.65

 laughing

0.65

 novice

0.64

 seed

0.64

Activations Density 0.059%

phrases related to changes or transitions

phrases indicating a change or transition in status or condition

No Comments

No Known Activations