INDEX

Explanations

phrases related to multiple instances or occurrences of something

oai_token-act-pair · gpt-3.5-turbo

references to the word "multiple."

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

naire

-0.68

adobe

-0.66

brance

-0.66

nothing

-0.65

instead

-0.64

ESE

-0.64

Dialogue

-0.63

emi

-0.62

NEY

-0.61

ACH

-0.60

POSITIVE LOGITS

 sclerosis

1.85

xes

1.68

 layers

1.18

 generations

1.14

 occasions

1.14

ples

1.14

 iterations

1.13

 versions

1.09

 viewpoints

1.09

 tiers

1.08

Activations Density 0.054%

phrases related to multiple instances or occurrences of something

references to the word "multiple."

No Comments

No Known Activations