INDEX

Explanations

words related to changing or transitioning

oai_token-act-pair · gpt-3.5-turbo

phrases related to changing states or transitions

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

EMENT

-0.72

INO

-0.68

EST

-0.66

LIB

-0.63

MU

-0.60

Spoiler

-0.59

UAL

-0.59

Brave

-0.59

 fullest

-0.58

POST

-0.57

POSITIVE LOGITS

imester

0.85

ipolar

0.84

ricular

0.78

izoph

0.77

apters

0.74

rients

0.74

levels

0.74

acco

0.72

rencies

0.71

drive

0.70

Activations Density 0.112%

words related to changing or transitioning

phrases related to changing states or transitions

No Comments

No Known Activations