INDEX

Explanations

phrases related to safety and security concerns, often in the context of government or law enforcement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 flesh

-0.67

 tubing

-0.61

 molten

-0.61

 muted

-0.60

ink

-0.60

 hose

-0.59

 boots

-0.58

 rampant

-0.58

 rake

-0.57

 sand

-0.57

POSITIVE LOGITS

 Strategies

1.03

 Relations

1.02

 Practices

1.01

 Computing

0.94

 Assessment

0.92

 Practice

0.92

 Resource

0.91

 Disorders

0.91

 Operations

0.90

 Profession

0.89

Activations Density 0.215%

phrases related to safety and security concerns, often in the context of government or law enforcement

No Comments

No Known Activations