INDEX

Explanations

references to coaching positions and related activities

oai_token-act-pair · gpt-3.5-turbo

terms related to coaching and instructional 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

hens

-0.83

ighed

-0.80

imi

-0.78

oning

-0.76

ifiers

-0.74

ewitness

-0.74

mington

-0.73

ivic

-0.72

ifier

-0.71

pard

-0.69

POSITIVE LOGITS

 prowess

1.28

 chops

1.13

 excellence

1.09

 styles

1.07

 skills

1.06

 talent

1.05

 expertise

1.04

 style

1.03

 talents

0.98

 ability

0.98

Activations Density 0.195%

references to coaching positions and related activities

terms related to coaching and instructional roles

No Comments

No Known Activations