INDEX

Explanations

locations or places ending with 'pur'

oai_token-act-pair · gpt-3.5-turbo

references to specific geographic locations or places

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

leg

-0.70

BIL

-0.69

llan

-0.64

MT

-0.64

ez

-0.63

WE

-0.62

GN

-0.62

uel

-0.62

asp

-0.61

yn

-0.61

POSITIVE LOGITS

etheless

0.92

ipal

0.79

ãĤ¡

0.78

theless

0.77

ocobo

0.74

obile

0.74

geons

0.72

ãĥĨãĤ£

0.72

vous

0.71

acters

0.69

Activations Density 0.051%

locations or places ending with 'pur'

references to specific geographic locations or places

No Comments

No Known Activations