INDEX

Explanations

words related to employment and workforce

oai_token-act-pair · gpt-3.5-turbo

suffixes related to professions and actions in occupational contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

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

 Haram

-0.64

channelAvailability

-0.62

combe

-0.62

 reper

-0.61

Dh

-0.56

 Sina

-0.55

Buk

-0.55

 Scotia

-0.54

INST

-0.54

XM

-0.52

POSITIVE LOGITS

ers

3.02

er

2.88

ership

2.21

ering

2.07

ERS

2.05

ER

2.00

ered

1.82

ery

1.65

erd

1.62

erness

1.62

Activations Density 0.340%

words related to employment and workforce

suffixes related to professions and actions in occupational contexts

No Comments

No Known Activations