INDEX

Explanations

the word "to" occurring before a certain context, often followed by an action or statement

oai_token-act-pair · gpt-3.5-turbo

the word "to" and its various uses in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

rieved

-0.70

pict

-0.61

ergy

-0.59

 prenatal

-0.59

job

-0.58

flight

-0.58

 plans

-0.57

 lines

-0.57

SpaceEngineers

-0.57

 reserves

-0.56

POSITIVE LOGITS

ilet

1.27

wered

1.12

ggles

1.11

 reiterate

1.08

 paraph

1.06

adies

0.99

 summarize

0.97

 compensate

0.94

asts

0.94

 illustrate

0.93

Activations Density 0.078%

the word "to" occurring before a certain context, often followed by an action or statement

the word "to" and its various uses in the text

No Comments

No Known Activations