INDEX

Explanations

mentions of excerpts

oai_token-act-pair · gpt-3.5-turbo

references to excerpts or sections of documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 guts

-0.71

 heights

-0.68

riter

-0.64

 Roose

-0.63

Kuh

-0.62

 crawl

-0.62

MENTS

-0.61

geist

-0.61

 straw

-0.61

taboola

-0.59

POSITIVE LOGITS

posure

1.11

haust

1.10

ternally

1.09

terior

1.06

change

1.05

odus

1.05

cerpt

1.01

ercise

1.01

asper

1.01

pert

1.01

Activations Density 0.010%

mentions of excerpts

references to excerpts or sections of documents

No Comments

No Known Activations