INDEX

Explanations

references to orphans and unrecognized entities

oai_token-act-pair · gpt-3.5-turbo

terms related to orphans and their situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

reens

-0.90

RT

-0.81

UD

-0.81

Hour

-0.77

ickr

-0.71

aeda

-0.71

andals

-0.69

rontal

-0.69

ordan

-0.68

rim

-0.68

POSITIVE LOGITS

 orphan

1.46

 orphans

1.29

phan

1.00

phans

0.89

sylvania

0.81

 Annie

0.81

 Icar

0.79

pup

0.76

minecraft

0.75

 pige

0.74

Activations Density 0.004%

references to orphans and unrecognized entities

terms related to orphans and their situations

No Comments

No Known Activations