INDEX

Explanations

instances where someone is stepping down or being forced out of a position in an organization

oai_token-act-pair · gpt-3.5-turbo

Triggered by @bot

references to job positions and roles within organizations

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 contrace

-0.77

hens

-0.76

ivers

-0.71

iven

-0.68

veland

-0.67

ulse

-0.67

apers

-0.66

aves

-0.63

rencies

-0.63

eches

-0.63

POSITIVE LOGITS

 position

1.57

 positions

1.55

job

1.43

 jobs

1.39

 posts

1.39

 roles

1.38

 role

1.38

 duties

1.37

 postings

1.23

 responsibilities

1.18

Activations Density 0.486%

instances where someone is stepping down or being forced out of a position in an organization

references to job positions and roles within organizations

No Known Activations