INDEX

Explanations

instances of regulatory or restrictive contexts, particularly in law enforcement scenarios

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

wcs

-1.18

luaj

-0.95

apy

-0.94

omore

-0.86

 ILCS

-0.86

akespeare

-0.85

ãĤ¨ãĥ«

-0.85

Ħ¢

-0.85

 gobl

-0.85

dimension

-0.84

POSITIVE LOGITS

 checkpoints

1.15

 checkpoint

1.01

hurst

0.74

 Hamas

0.73

arettes

0.69

 Ferry

0.69

ndum

0.67

 Israelis

0.67

 detained

0.67

 Palestinians

0.66

Activations Density 0.373%

instances of regulatory or restrictive contexts, particularly in law enforcement scenarios

No Comments

No Known Activations