INDEX

Explanations

words related to rims

oai_token-act-pair · gpt-3.5-turbo

references to the term "rim."

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

anders

-0.74

herty

-0.68

ettings

-0.64

IVES

-0.63

asma

-0.62

essions

-0.62

 confir

-0.60

 lawy

-0.59

 Idle

-0.58

Interstitial

-0.58

POSITIVE LOGITS

senal

1.33

med

1.12

ming

1.11

ned

0.95

eless

0.93

eworks

0.89

Rim

0.83

neys

0.83

phal

0.81

ply

0.80

Activations Density 0.025%

words related to rims

references to the term "rim."

No Comments

No Known Activations