INDEX

Explanations

mentions of formal discussions or negotiations between parties

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "talks" in various contexts

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

CTV

-0.79

Lake

-0.70

negie

-0.69

Merit

-0.67

rehens

-0.65

aples

-0.65

served

-0.64

Sensor

-0.64

rons

-0.63

IELD

-0.62

POSITIVE LOGITS

 Talks

0.89

 talks

0.84

hran

0.83

 negotiations

0.81

pace

0.80

how

0.79

afety

0.74

peed

0.74

paces

0.73

 talk

0.72

Activations Density 0.017%

mentions of formal discussions or negotiations between parties

occurrences of the word "talks" in various contexts

No Comments

No Known Activations