INDEX

Explanations

words related to sports coaches

oai_token-act-pair · gpt-3.5-turbo

references to coaches in various sports contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Diss

-0.72

anium

-0.68

objects

-0.68

ÅĤ

-0.65

dor

-0.64

 digit

-0.62

ases

-0.61

 Resistance

-0.61

ashed

-0.61

 Render

-0.60

POSITIVE LOGITS

 coach

3.69

 Coach

2.92

Coach

2.86

 coaches

2.78

 coaching

2.22

 coached

1.92

 trainer

1.74

 teammate

1.64

 tutor

1.51

 manager

1.45

Activations Density 0.028%

words related to sports coaches

references to coaches in various sports contexts

No Comments

No Known Activations