INDEX

Explanations

numerical values associated with large quantities

oai_token-act-pair · gpt-3.5-turbo

phrases indicating large quantities or figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ffen

-0.62

":[

-0.62

ittal

-0.62

ILY

-0.59

illary

-0.57

 Correction

-0.56

egal

-0.55

 Dynamics

-0.54

 Eventually

-0.52

arten

-0.52

POSITIVE LOGITS

drive

1.23

capacity

1.11

priced

1.09

kill

1.03

loading

1.02

valued

0.99

whelming

0.99

grown

0.98

hang

0.97

Activations Density 0.058%

numerical values associated with large quantities

phrases indicating large quantities or figures

No Comments

No Known Activations