INDEX

Explanations

words with the sub-string "pl"

oai_token-act-pair · gpt-3.5-turbo

instances of the prefix "pl-" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

âĸ¬

-0.87

PRES

-0.80

âĸ¬âĸ¬

-0.79

BOOK

-0.79

VICE

-0.75

hyde

-0.74

Madison

-0.73

FINE

-0.70

FINEST

-0.68

 Apprentice

-0.68

POSITIVE LOGITS

atter

1.17

ump

1.16

umber

1.16

aques

1.14

umbers

1.11

umb

1.07

acent

1.06

asma

1.05

ipp

1.05

acement

1.03

Activations Density 0.013%

words with the sub-string "pl"

instances of the prefix "pl-" in various contexts

No Comments

No Known Activations