INDEX

Explanations

references to a particular TV series or show

oai_token-act-pair · gpt-3.5-turbo

mentions of series or shows, particularly regarding new or upcoming productions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

zie

-0.87

orget

-0.79

attled

-0.77

lesi

-0.72

artisan

-0.69

lig

-0.68

vae

-0.67

 saline

-0.66

aths

-0.65

rian

-0.65

POSITIVE LOGITS

 finale

0.90

manship

0.84

enegger

0.78

 Reborn

0.75

 regulars

0.75

ĸļ

0.70

arium

0.69

 adaptation

0.66

peak

0.66

hews

0.66

Activations Density 0.025%

references to a particular TV series or show

mentions of series or shows, particularly regarding new or upcoming productions

No Comments

No Known Activations