INDEX

Explanations

references to wide receiver positions in sports contexts

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

anamo

-0.76

omorphic

-0.76

 Debor

-0.72

onyms

-0.69

 Bulgar

-0.67

 Maid

-0.65

cles

-0.63

arial

-0.61

 Azerbai

-0.59

phy

-0.59

POSITIVE LOGITS

ONG

1.34

ACK

1.08

MJ

1.02

EST

1.01

ITT

0.99

IGHT

0.97

ATH

0.96

IVES

0.91

IGHTS

0.87

WR

0.87

Activations Density 0.005%

references to wide receiver positions in sports contexts

No Comments

No Known Activations