INDEX

Explanations

mentions of specific locations or organizations, particularly related to serious events or history

oai_token-act-pair · gpt-3.5-turbo

mentions of specific geographical locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Entity

-0.73

VALUE

-0.70

TPPStreamerBot

-0.67

IPS

-0.67

PATH

-0.67

Ent

-0.66

Tokens

-0.66

 NASCAR

-0.65

ECT

-0.65

INT

-0.65

POSITIVE LOGITS

reb

1.35

acus

0.87

unal

0.86

vernment

0.84

oli

0.83

aternity

0.82

undy

0.81

axter

0.77

ruary

0.77

iotics

0.75

Activations Density 0.020%

mentions of specific locations or organizations, particularly related to serious events or history

mentions of specific geographical locations

No Comments

No Known Activations