INDEX

Explanations

references to specific episodes in a TV series

oai_token-act-pair · gpt-3.5-turbo

mentions of specific TV show episodes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

sheets

-0.86

minist

-0.84

ndra

-0.78

sheet

-0.78

press

-0.77

 wheelchair

-0.74

 denomin

-0.74

folk

-0.73

gling

-0.72

enei

-0.72

POSITIVE LOGITS

 Episode

1.12

Episode

1.00

 Transcript

0.90

 episode

0.82

isodes

0.82

episode

0.81

 Explicit

0.78

 Ratings

0.74

 Tickets

0.74

 episodes

0.73

Activations Density 0.009%

references to specific episodes in a TV series

mentions of specific TV show episodes

No Comments

No Known Activations