INDEX

Explanations

people or things referred to as "agents."

oai_token-act-pair · gpt-3.5-turbo

references to various types of agents, particularly in contexts involving espionage or undercover operations

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

issance

-0.86

FACE

-0.84

lihood

-0.83

etooth

-0.73

jri

-0.72

ths

-0.71

rise

-0.69

vironment

-0.67

frey

-0.67

cept

-0.67

POSITIVE LOGITS

 prov

1.12

 agent

0.94

 posing

0.92

 Agent

0.91

 Agents

0.90

 agents

0.90

agent

0.90

Agent

0.81

agents

0.80

iola

0.76

Activations Density 0.039%

people or things referred to as "agents."

references to various types of agents, particularly in contexts involving espionage or undercover operations

No Comments

No Known Activations