INDEX

Explanations

terms related to removing or reducing certain elements or qualities from something

oai_token-act-pair · gpt-3.5-turbo

words related to string manipulation or structures

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

layer

-0.70

nu

-0.67

icum

-0.65

 Leilan

-0.65

 Aval

-0.64

amd

-0.63

 aval

-0.62

lan

-0.61

 Suffolk

-0.61

 Messenger

-0.60

POSITIVE LOGITS

pped

1.80

pping

1.74

ppers

1.51

pper

1.50

ppy

1.46

ppings

1.40

ps

1.39

zzle

1.35

ggle

1.32

ker

1.29

Activations Density 0.058%

terms related to removing or reducing certain elements or qualities from something

words related to string manipulation or structures

No Comments

No Known Activations