INDEX

Explanations

names, particularly first and last names

oai_token-act-pair · gpt-3.5-turbo

references to specific names or terms associated with individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 cytok

-0.69

 destro

-0.68

pen

-0.67

 stocking

-0.67

 chim

-0.66

senal

-0.66

 inval

-0.64

 repe

-0.63

rers

-0.63

 volt

-0.63

POSITIVE LOGITS

\\\\\\\\

0.97

uben

0.86

nesses

0.82

esley

0.79

eus

0.77

ica

0.77

heid

0.76

ITIES

0.74

orr

0.74

elfth

0.73

Activations Density 0.019%

names, particularly first and last names

references to specific names or terms associated with individuals

No Comments

No Known Activations