INDEX

Explanations

instances of hiring or employment-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-att_32k-oai

Configuration

jbloom/GPT2-Small-OAI-v5-32k-attn-out-SAEs/v5_32k_layer_3

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

32,768

Data Type

torch.float32

Hook Name

blocks.3.hook_attn_out

Hook Layer

Architecture

standard

Context Size

Dataset

Skylion007/openwebtext

Activation Function

topk

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĸļ

-0.70

phy

-0.68

ipel

-0.67

agy

-0.66

 choking

-0.64

ixel

-0.64

pha

-0.64

limits

-0.63

nown

-0.63

icent

-0.63

POSITIVE LOGITS

Deal

0.82

FML

0.70

inance

0.68

 payroll

0.66

Dream

0.66

 bait

0.66

ItemTracker

0.65

NRS

0.62

EngineDebug

0.61

 intel

0.59

Activations Density 0.107%

instances of hiring or employment-related terms

No Comments

No Known Activations