INDEX

Explanations

occurrences or examples of specific situations or events

oai_token-act-pair · gpt-3.5-turbo

mentions of various instances or occurrences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ramid

-0.85

inally

-0.77

eday

-0.77

price

-0.71

sett

-0.70

jri

-0.70

bern

-0.69

isoft

-0.68

yss

-0.68

mosp

-0.67

POSITIVE LOGITS

 instances

1.00

uations

0.95

 occurrences

0.76

 cases

0.71

 occurrence

0.71

wcsstore

0.69

 thereof

0.67

manship

0.66

 Flask

0.64

uation

0.64

Activations Density 0.009%

occurrences or examples of specific situations or events

mentions of various instances or occurrences

No Comments

No Known Activations