INDEX

Explanations

locations and organizations

oai_token-act-pair · gpt-3.5-turbo

commas used in lists or separate clauses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

epad

-0.72

eters

-0.65

atics

-0.62

atson

-0.62

ubes

-0.61

herty

-0.61

efe

-0.60

etermin

-0.59

ambers

-0.58

ocry

-0.58

POSITIVE LOGITS

aka

1.06

Inc

0.99

LLC

0.98

 meanwhile

0.96

huh

0.91

 formerly

0.84

 supra

0.77

 birthplace

0.76

INC

0.74

 Volume

0.74

Activations Density 0.183%

locations and organizations

commas used in lists or separate clauses

No Comments

No Known Activations