INDEX

Explanations

references to job-related terms and employment issues

oai_token-act-pair · gpt-3.5-turbo

terms related to employment and job statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Oo

-0.76

issance

-0.66

anwhile

-0.65

++++

-0.63

ateur

-0.62

åħī

-0.62

adoes

-0.62

ideshow

-0.62

Riy

-0.62

ATURE

-0.61

POSITIVE LOGITS

sworth

0.81

anooga

0.79

hunt

0.78

Lov

0.77

seekers

0.77

ulia

0.77

fell

0.77

 Maker

0.75

Job

0.74

lins

0.74

Activations Density 0.030%

references to job-related terms and employment issues

terms related to employment and job statistics

No Comments

No Known Activations