INDEX

Explanations

instances of the word "am" followed by another word

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "am."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Clause

-0.70

 aver

-0.65

 concession

-0.62

 wart

-0.62

 legality

-0.61

 Durham

-0.61

 discretionary

-0.61

 uranium

-0.59

 takeaway

-0.58

 document

-0.58

POSITIVE LOGITS

ateurs

1.45

ethyst

1.39

azon

1.38

nesia

1.26

icable

1.26

assing

1.24

enable

1.19

icably

1.14

azes

1.12

azing

1.12

Activations Density 0.020%

instances of the word "am" followed by another word

occurrences of the word "am."

No Comments

No Known Activations