INDEX

Explanations

names, abbreviations, and terms related to organizations or entities

oai_token-act-pair · gpt-3.5-turbo

specific acronyms, names, and complex identifiers related to organizations or individuals

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

taboola

-1.22

ioxide

-0.67

BILITIES

-0.65

degree

-0.64

ciating

-0.62

iture

-0.60

lishing

-0.60

 guiActiveUnfocused

-0.60

webkit

-0.60

pse

-0.58

POSITIVE LOGITS

inki

1.13

igans

0.81

puff

0.74

sap

0.73

idan

0.73

 Abedin

0.71

quarters

0.71

ournals

0.71

ĸļ

0.70

nian

0.70

Activations Density 0.365%

names, abbreviations, and terms related to organizations or entities

specific acronyms, names, and complex identifiers related to organizations or individuals

No Known Activations