INDEX

Explanations

terms related to physical structures or locations, specifically focusing on houses

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 @ 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

aneously

-0.75

ICAL

-0.68

uyomi

-0.66

CNS

-0.66

 guiActiveUn

-0.65

istical

-0.65

ï¸ı

-0.64

NetMessage

-0.64

 clipboard

-0.63

 Rounds

-0.62

POSITIVE LOGITS

wife

1.61

wives

1.60

keeping

1.39

warming

1.16

keeper

1.14

plant

1.10

mates

1.09

keepers

1.05

boat

1.04

boats

1.03

Activations Density 0.051%

terms related to physical structures or locations, specifically focusing on houses

references to "house" in various contexts

No Comments

No Known Activations