INDEX

Explanations

language indicating a specific time or point in time

oai_token-act-pair · gpt-3.5-turbo

instances of a specific phrase or structure indicating a point in time

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

 punishable

-0.59

 favour

-0.58

itivity

-0.57

tools

-0.56

 entit

-0.56

 prosecut

-0.54

 favor

-0.53

forth

-0.53

direction

-0.51

 compl

-0.51

POSITIVE LOGITS

mosp

1.30

hens

1.27

 least

1.20

kinson

1.09

letico

1.04

omic

0.99

 stake

0.97

 present

0.92

roc

0.88

ivan

0.84

Activations Density 0.069%

language indicating a specific time or point in time

instances of a specific phrase or structure indicating a point in time

No Comments

No Known Activations