INDEX

Explanations

references to specific names, particularly individuals with the name "Nicholas" and "Christina."

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

doors

-0.96

itar

-0.88

æ©

-0.85

por

-0.84

olding

-0.83

rop

-0.82

rim

-0.77

atures

-0.76

ebook

-0.75

ARE

-0.74

POSITIVE LOGITS

vous

0.80

 carbohyd

0.74

ĪĴ

0.73

 manifests

0.73

etheless

0.73

 flared

0.72

ItemTracker

0.71

 hiatus

0.69

 occas

0.67

aughs

0.65

Activations Density 0.023%

references to specific names, particularly individuals with the name "Nicholas" and "Christina."

No Comments

No Known Activations