INDEX

Explanations

references to the University of California, specifically focusing on the Berkeley campus

oai_token-act-pair · gpt-3.5-turbo

references to the University of California (UC) system and its campuses

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

igree

-0.66

htar

-0.66

bringer

-0.65

 maiden

-0.65

 chained

-0.63

uese

-0.61

 undone

-0.61

esthetic

-0.61

 alternatively

-0.60

 crate

-0.60

POSITIVE LOGITS

 Berkeley

1.27

 Irvine

1.20

SC

1.06

 campuses

1.04

PD

1.04

MJ

1.02

 Riverside

1.02

UC

1.02

CLA

0.96

 Davis

0.93

Activations Density 0.024%

references to the University of California, specifically focusing on the Berkeley campus

references to the University of California (UC) system and its campuses

No Comments

No Known Activations