INDEX

Explanations

expressions related to critique or evaluation

oai_token-act-pair · gpt-3.5-turbo

phrases indicating superiority or excellence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Emergency

-0.68

pleted

-0.63

Located

-0.63

ospital

-0.59

icipated

-0.57

angering

-0.57

TBA

-0.56

displayText

-0.55

eligible

-0.55

 mosqu

-0.55

POSITIVE LOGITS

 analogy

1.02

 Nietzsche

0.80

 philosophers

0.80

 paraph

0.78

 theorists

0.75

 Chomsky

0.75

 admittedly

0.75

 quote

0.75

 irony

0.72

quote

0.71

Activations Density 0.960%

expressions related to critique or evaluation

phrases indicating superiority or excellence

No Comments

No Known Activations