INDEX

Explanations

locations or facilities that provide services or activities

oai_token-act-pair · gpt-3.5-turbo

terms related to facilities, services, and infrastructure

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

Beg

-0.69

ãĥĥãĥī

-0.66

oret

-0.64

PsyNetMessage

-0.63

patient

-0.63

ãĥ£

-0.62

 Aval

-0.61

abase

-0.59

 Hemisphere

-0.57

 Kaplan

-0.57

POSITIVE LOGITS

 alike

1.75

 respectively

1.49

 thereof

1.08

hips

0.99

etc

0.89

 attRot

0.89

 accordingly

0.84

 therein

0.84

 ensued

0.78

 thereto

0.77

Activations Density 0.560%

locations or facilities that provide services or activities

terms related to facilities, services, and infrastructure

No Comments

No Known Activations