INDEX

Explanations

phrases involving the word "id"

oai_token-act-pair · gpt-3.5-turbo

variations of the word "rid."

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

bilt

-0.67

andise

-0.66

 silence

-0.65

izabeth

-0.59

 Sieg

-0.59

 Winc

-0.57

 Bolton

-0.56

Nun

-0.56

consider

-0.55

 outlets

-0.55

POSITIVE LOGITS

irection

1.29

ividual

1.15

iots

1.13

ocious

1.12

eways

1.07

iosyncr

1.04

gets

1.04

imensional

1.03

ocument

1.01

uci

1.00

Activations Density 0.049%

phrases involving the word "id"

variations of the word "rid."

No Comments

No Known Activations