INDEX

Explanations

phrases related to sports teams and events

oai_token-act-pair · gpt-3.5-turbo

occurrences of specific characters or letters 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 @ 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

 refrain

-0.64

stellar

-0.63

 retract

-0.62

pac

-0.61

 Feder

-0.61

 transitional

-0.60

flu

-0.59

 upfront

-0.59

 Trin

-0.59

 reception

-0.58

POSITIVE LOGITS

ageddon

0.96

akery

0.93

ackers

0.91

akers

0.91

acker

0.90

utters

0.89

ights

0.88

angers

0.87

igs

0.86

atcher

0.86

Activations Density 0.110%

phrases related to sports teams and events

occurrences of specific characters or letters in the text

No Comments

No Known Activations