INDEX

Explanations

text excerpts

oai_token-act-pair · gpt-3.5-turbo

terms related to excerpts and passages from various texts

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

ggle

-0.85

thood

-0.82

vil

-0.76

gaard

-0.67

kos

-0.66

ces

-0.65

ggles

-0.64

 vacant

-0.64

mos

-0.63

opter

-0.62

POSITIVE LOGITS

 excerpts

1.28

 excerpt

1.08

 guiActiveUn

1.03

cerpt

1.01

DragonMagazine

1.01

 snippets

0.94

 extracts

0.86

="#

0.79

icularly

0.78

...]

0.76

Activations Density 0.011%

text excerpts

terms related to excerpts and passages from various texts

No Comments

No Known Activations