INDEX

Explanations

positions in sports

oai_token-act-pair · gpt-3.5-turbo

terms related to football player positions and roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.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

ombat

-0.89

pload

-0.73

 Newsp

-0.69

 Roose

-0.67

 Humanity

-0.66

 Agenda

-0.65

llah

-0.64

 Speech

-0.63

DATA

-0.63

sth

-0.63

POSITIVE LOGITS

 tandem

1.02

 extraord

1.01

 corps

1.00

/.

0.95

 coach

0.91

 drafted

0.89

duo

0.85

 prospect

0.85

 position

0.83

 pairing

0.83

Activations Density 0.109%

positions in sports

terms related to football player positions and roles

No Comments

No Known Activations