INDEX

Explanations

mathematical or technical jargon related to data processing or analysis

oai_token-act-pair · gpt-3.5-turbo

forms of the verb "to be."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Newspaper

-0.73

 Pear

-0.64

lights

-0.62

 Newsp

-0.61

 Need

-0.61

 Coff

-0.61

 Guilty

-0.58

 Lauder

-0.58

pour

-0.57

Rab

-0.57

POSITIVE LOGITS

 traced

1.25

 summed

1.05

 forgiven

1.04

 likened

1.02

 manipulated

1.01

 achieved

1.01

 summarized

1.00

 safely

0.98

 accessed

0.98

 avoided

0.97

Activations Density 0.078%

mathematical or technical jargon related to data processing or analysis

forms of the verb "to be."

No Comments

No Known Activations