INDEX

Explanations

names of people or locations

oai_token-act-pair · gpt-3.5-turbo

names and terms related to individuals, particularly in a legal or news context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

•EggList

No Comments

Negative Logits

interstitial

-0.84

 intermedi

-0.83

INF

-0.81

 Malone

-0.73

-0.72

 Instr

-0.71

agascar

-0.69

 Alexis

-0.68

-0.67

POSITIVE LOGITS

1.24

1.15

WM

1.12

wi

1.08

wit

1.05

WC

1.03

wl

1.02

wa

1.02

wt

1.02

wal

1.02

Activations Density 0.208%

names of people or locations

names and terms related to individuals, particularly in a legal or news context

No Comments

No Known Activations