INDEX

Explanations

instances of the word "tamper" and its variations, indicating a focus on tampering-related activities or events

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

Ĥª

-0.73

 Prospect

-0.71

 Dayton

-0.69

SHIP

-0.67

 Johann

-0.66

stood

-0.64

 adequ

-0.63

 expectancy

-0.63

 Spectrum

-0.62

 Borders

-0.60

POSITIVE LOGITS

pered

1.27

tam

1.13

pering

1.04

eless

1.00

inated

0.99

ming

0.90

arella

0.89

iour

0.86

inating

0.86

iny

0.85

Activations Density 0.005%

instances of the word "tamper" and its variations, indicating a focus on tampering-related activities or events

No Comments

No Known Activations