INDEX

Explanations

instances of the word "house."

oai_token-act-pair · gpt-3.5-turbo

references to "house" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

istically

-0.75

 Archdemon

-0.70

umbo

-0.68

DPR

-0.67

PsyNetMessage

-0.67

*/(

-0.66

 timelines

-0.63

uyomi

-0.62

 periphery

-0.61

 synd

-0.61

POSITIVE LOGITS

hold

1.22

door

1.10

wife

1.10

wives

1.04

warming

1.01

holder

0.91

holders

0.91

keeper

0.90

keeping

0.90

cle

0.89

Activations Density 0.012%

instances of the word "house."

references to "house" in various contexts

No Comments

No Known Activations