INDEX

Explanations

references to the CNN news network

oai_token-act-pair · gpt-3.5-turbo

mentions of CNN

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Baal

-0.69

plates

-0.68

 monument

-0.67

pell

-0.66

phant

-0.64

 blinded

-0.63

tons

-0.62

piring

-0.62

 monuments

-0.61

ocene

-0.60

POSITIVE LOGITS

Money

1.45

pedia

1.01

 affiliate

0.99

NOW

0.95

iliate

0.89

 Money

0.89

igel

0.86

 Tonight

0.86

CNN

0.86

 anchor

0.85

Activations Density 0.022%

references to the CNN news network

mentions of CNN

No Comments

No Known Activations