INDEX

Explanations

phrases or sentences describing initial actions or stages in a process

oai_token-act-pair · gpt-3.5-turbo

references to "first steps" or initial actions 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 @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ores

-0.75

iqueness

-0.69

 Sins

-0.69

olls

-0.69

inately

-0.67

anguages

-0.66

 contracted

-0.66

licted

-0.66

ittens

-0.66

poons

-0.65

POSITIVE LOGITS

daughter

0.99

 Steps

0.94

step

0.91

 step

0.91

Step

0.91

 steps

0.90

steps

0.86

 toward

0.85

dad

0.84

 forward

0.81

Activations Density 0.018%

phrases or sentences describing initial actions or stages in a process

references to "first steps" or initial actions in various contexts

No Comments

No Known Activations