INDEX

Explanations

proper nouns related to individuals named Wendy

oai_token-act-pair · gpt-3.5-turbo

references to the name "Wendy" and related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

dinand

-1.01

ertodd

-0.86

hari

-0.83

raltar

-0.82

sburg

-0.80

estones

-0.77

estone

-0.76

ularity

-0.75

rawdownloadcloneembedreportprint

-0.75

ember

-0.75

POSITIVE LOGITS

Bee

0.84

 Wendy

0.73

igmatic

0.65

ryu

0.60

roid

0.60

lock

0.60

bun

0.60

 glove

0.60

Umb

0.60

 squeeze

0.59

Activations Density 0.029%

proper nouns related to individuals named Wendy

references to the name "Wendy" and related concepts

No Comments

No Known Activations