INDEX

Explanations

proper nouns related to locations or religious figures

oai_token-act-pair · gpt-3.5-turbo

references to saints and their associated names or locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

hower

-0.70

 timely

-0.66

ittee

-0.61

swing

-0.61

clips

-0.60

OTH

-0.60

ESE

-0.60

Ĥİ

-0.59

qua

-0.58

å¥³

-0.58

POSITIVE LOGITS

 Lucia

1.23

 Augustine

1.17

 Kitt

1.15

 Petersburg

1.13

 Vincent

1.09

 Helena

1.09

 Jude

1.08

 Barth

1.07

 Dominic

1.01

 Benedict

0.99

Activations Density 0.035%

proper nouns related to locations or religious figures

references to saints and their associated names or locations

No Comments

No Known Activations