INDEX

Explanations

terms related to academic institutions, particularly schools and departments within universities

oai_token-act-pair · gpt-3.5-turbo

mentions of academic institutions and associated programs

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

 selves

-0.57

 ensu

-0.55

numbered

-0.55

 wagon

-0.52

tee

-0.52

"]=>

-0.51

zers

-0.50

 bleed

-0.49

 lockout

-0.49

++++

-0.49

POSITIVE LOGITS

 Research

0.78

Bio

0.72

 Institute

0.70

 Archae

0.66

 Laboratories

0.62

 University

0.61

 Laboratory

0.61

Research

0.60

 Economic

0.60

 Fell

0.60

Activations Density 0.998%

terms related to academic institutions, particularly schools and departments within universities

mentions of academic institutions and associated programs

No Comments

No Known Activations