INDEX

Explanations

pronouns indicating unidentified entities or situations

oai_token-act-pair · gpt-3.5-turbo

the pronoun "it"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

DAQ

-0.76

ROR

-0.64

ZA

-0.62

IVE

-0.58

IONS

-0.56

pmwiki

-0.55

natureconservancy

-0.54

EMBER

-0.54

UG

-0.54

ALE

-0.53

POSITIVE LOGITS

it

2.01

it

1.04

It

1.04

It

1.00

its

0.95

 there

0.86

 they

0.85

 this

0.84

 everything

0.80

you

0.78

Activations Density 0.319%

pronouns indicating unidentified entities or situations

the pronoun "it"

No Comments

No Known Activations