INDEX

Explanations

phrases emphasizing uniqueness or distinction

oai_token-act-pair · gpt-3.5-turbo

expressions emphasizing uniqueness or singularity

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 cannabin

-0.82

idates

-0.76

kefeller

-0.69

istries

-0.64

iculty

-0.61

rity

-0.60

incial

-0.60

idity

-0.59

uliffe

-0.59

osterone

-0.59

POSITIVE LOGITS

CLASSIFIED

0.66

 plague

0.66

ippers

0.64

 magnets

0.64

mother

0.63

 Honey

0.62

devices

0.62

ahime

0.61

 Ancients

0.61

hots

0.60

Activations Density 0.128%

phrases emphasizing uniqueness or distinction

expressions emphasizing uniqueness or singularity

No Comments

No Known Activations