INDEX

Explanations

references to specific individuals or characters

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

WARD

-0.74

aughs

-0.72

))))

-0.67

idian

-0.66

delay

-0.63

 Bezos

-0.63

}}

-0.63

 VIDEOS

-0.62

indu

-0.62

actual

-0.61

POSITIVE LOGITS

amera

0.73

 largeDownload

0.71

 Made

0.67

nutrition

0.63

 ranking

0.63

qualified

0.63

pid

0.62

 Miko

0.60

sted

0.59

 Cipher

0.58

Activations Density 3.517%

references to specific individuals or characters

No Comments

No Known Activations