INDEX

Explanations

references to the television network PBS

oai_token-act-pair · gpt-3.5-turbo

mentions of television broadcasting networks and related programming

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nesota

-0.66

charged

-0.64

ometry

-0.60

hunt

-0.60

headers

-0.60

 finding

-0.60

butt

-0.59

ning

-0.59

////////////////////////////////

-0.59

liners

-0.59

POSITIVE LOGITS

DAQ

0.86

PBS

0.84

FIL

0.80

kies

0.78

 Interstellar

0.75

 Studios

0.74

nown

0.73

LV

0.72

 Films

0.72

ACE

0.72

Activations Density 0.007%

references to the television network PBS

mentions of television broadcasting networks and related programming

No Comments

No Known Activations