INDEX

Explanations

phrases related to professional positions and roles

oai_token-act-pair · gpt-3.5-turbo

phrases related to professional roles and affiliations

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

Instead

-0.70

License

-0.67

LESS

-0.67

iru

-0.66

0000000000000000

-0.65

Asked

-0.65

pmwiki

-0.65

acs

-0.64

 surpr

-0.63

rained

-0.61

POSITIVE LOGITS

 numerous

0.99

 several

0.96

 Presidents

0.91

 various

0.90

 both

0.87

 countless

0.78

 stint

0.77

ĸļ

0.75

 Seasons

0.73

 seminal

0.73

Activations Density 0.313%

phrases related to professional positions and roles

phrases related to professional roles and affiliations

No Comments

No Known Activations