INDEX

Explanations

monetary amounts or measurements like dollars, numbers, or units of measurement

oai_token-act-pair · gpt-3.5-turbo

abbreviations or acronyms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Franz

-0.82

 advis

-0.80

 Indra

-0.77

Jas

-0.76

 Sins

-0.73

 Strauss

-0.72

 enlarg

-0.72

 wings

-0.71

ãĥł

-0.71

 Mysterious

-0.70

POSITIVE LOGITS

et

1.56

ets

1.34

ET

1.34

Ã©t

1.29

ett

1.24

ete

1.21

etry

1.10

ETS

1.08

etta

1.05

etric

1.05

Activations Density 0.202%

monetary amounts or measurements like dollars, numbers, or units of measurement

abbreviations or acronyms

No Comments

No Known Activations