INDEX

Explanations

phrases pertaining to qualifications or expertise

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

Not in Any Lists

No Comments

Negative Logits

âĸ¬

-0.82

 Gardens

-0.75

 Lans

-0.72

 Spice

-0.72

 drip

-0.70

 DRAG

-0.67

 Tycoon

-0.66

 Sparrow

-0.66

 Torn

-0.65

 Flow

-0.64

POSITIVE LOGITS

 qualifications

0.95

animous

0.90

 applicants

0.88

itatively

0.88

dinand

0.86

IFIED

0.84

ifications

0.83

qualified

0.82

ifiers

0.82

ified

0.81

Activations Density 0.006%

phrases pertaining to qualifications or expertise

No Comments

No Known Activations