INDEX

Explanations

words containing the sequence "ron" followed by a number indicating significance

oai_token-act-pair · gpt-3.5-turbo

occurrences of the token "ron," indicating a focus on names or terms ending with "ron."

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

tub

-0.79

urai

-0.69

PROV

-0.64

TPS

-0.63

RED

-0.63

 Virtue

-0.62

URA

-0.61

 Jackets

-0.61

AY

-0.60

 Norn

-0.59

POSITIVE LOGITS

 Collider

1.09

auts

1.04

aldo

1.03

autical

0.99

omical

0.96

omics

0.95

ising

0.92

stadt

0.88

uclear

0.88

naire

0.87

Activations Density 0.033%

words containing the sequence "ron" followed by a number indicating significance

occurrences of the token "ron," indicating a focus on names or terms ending with "ron."

No Comments

No Known Activations