INDEX

Explanations

words related to topography or positional terms

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "top" 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 discord

-0.64

uctions

-0.63

 labour

-0.62

 Gaul

-0.62

 Consent

-0.61

 Baldwin

-0.61

 vomit

-0.60

 Kath

-0.60

 hunger

-0.60

 consent

-0.59

POSITIVE LOGITS

most

1.10

ographical

1.10

ography

1.01

ographically

0.96

deck

0.95

ology

0.93

ographic

0.88

ologically

0.86

eka

0.83

mast

0.82

Activations Density 0.035%

words related to topography or positional terms

references to the concept of "top" in various contexts

No Comments

No Known Activations