INDEX

Explanations

phrases reflecting legal issues and statements regarding accountability

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

IFrame

Link

Negative Logits

 misunder

-0.71

 visas

-0.67

 hitters

-0.67

 outputs

-0.67

 necessities

-0.66

 mistakes

-0.64

 machines

-0.63

 targets

-0.63

 bodies

-0.63

 spirits

-0.62

POSITIVE LOGITS

Jr

1.03

Sr

0.99

QC

0.87

 Founder

0.86

 Managing

0.86

Jr

0.84

 Chair

0.83

 Director

0.82

aka

0.80

 Publisher

0.80

Activations Density 0.077%

phrases reflecting legal issues and statements regarding accountability

No Known Activations