INDEX

Explanations

words related to a specific writing technique or prompt

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

New Auto-Interp

Configuration

jbloom/GPT2-Small-OAI-v5-128k-attn-out-SAEs/v5_128k_layer_0

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

131,072

Data Type

torch.float32

Hook Name

blocks.0.hook_attn_out

Hook Layer

Architecture

standard

Context Size

Dataset

Skylion007/openwebtext

Activation Function

topk

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

genic

-0.82

Leaks

-0.78

bably

-0.78

 doping

-0.71

 corrid

-0.71

 withdrawing

-0.70

uner

-0.68

 leakage

-0.68

 undet

-0.68

cess

-0.67

POSITIVE LOGITS

uyomi

0.84

 Dome

0.80

uliffe

0.80

sonian

0.80

racuse

0.78

*/(

0.77

 Bastard

0.76

 Higgins

0.73

 Dispatch

0.73

 Concord

0.72

Activations Density 0.046%

words related to a specific writing technique or prompt

No Comments

No Known Activations