INDEX

Explanations

words related to living environments or locations

oai_token-act-pair · gpt-3.5-turbo

mentions of living or dwelling places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gone

-0.76

reg

-0.73

guided

-0.72

gotten

-0.70

umbs

-0.70

displayText

-0.67

imb

-0.67

Cola

-0.66

 damned

-0.66

err

-0.64

POSITIVE LOGITS

 residence

1.06

 halls

0.92

chool

0.83

maid

0.83

coming

0.82

loo

0.82

iage

0.80

 Halls

0.79

 residences

0.78

 hall

0.77

Activations Density 0.008%

words related to living environments or locations

mentions of living or dwelling places

No Comments

No Known Activations