INDEX

Explanations

video-related keywords and the word "WATCH"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "WATCH" followed by related video content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ctrl

-0.84

ifice

-0.66

phi

-0.65

ality

-0.65

ãĤ¨ãĥ«

-0.65

 misunderstanding

-0.65

 reopen

-0.64

vet

-0.63

 retard

-0.63

VEN

-0.63

POSITIVE LOGITS

 Watching

1.27

tower

1.03

Watch

1.01

 Watch

0.98

dog

0.94

 WATCH

0.90

watch

0.90

 Below

0.86

 Dogs

0.85

 Trend

0.81

Activations Density 0.014%

video-related keywords and the word "WATCH"

occurrences of the word "WATCH" followed by related video content

No Comments

No Known Activations