INDEX

Explanations

mentions of specific people, places, or significant events

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

ModLoader

-0.77

eneg

-0.72

NetMessage

-0.71

pmwiki

-0.68

Write

-0.65

weekly

-0.63

wiki

-0.63

 Offline

-0.61

 unethical

-0.61

 geographically

-0.60

POSITIVE LOGITS

 Tanz

0.70

OHN

0.68

sic

0.66

 Instruction

0.66

 whence

0.64

 Chim

0.64

Benz

0.63

 Coliseum

0.63

Ń·

0.62

Caf

0.61

Activations Density 0.076%

mentions of specific people, places, or significant events

No Comments

No Known Activations