INDEX

Explanations

phrases related to following or subsequent events

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Following" and its variations

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

enf

-0.80

uci

-0.79

Ñĭ

-0.76

cia

-0.75

adj

-0.74

mpire

-0.74

å¥³

-0.74

educ

-0.72

çļ

-0.70

ensity

-0.70

POSITIVE LOGITS

 revelations

0.95

 suit

0.93

 closure

0.90

 completion

0.89

 graduation

0.89

 weeks

0.85

 complaints

0.84

 extensive

0.83

 lengthy

0.83

 outcry

0.80

Activations Density 0.047%

phrases related to following or subsequent events

instances of the word "Following" and its variations

No Comments

No Known Activations