INDEX

Explanations

references to power capacity measurements in gigawatts (GW)

oai_token-act-pair · gpt-3.5-turbo

mentions of specific geographical locations and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

iasis

-0.75

ĺħ

-0.73

cision

-0.72

phrine

-0.70

netflix

-0.70

hyde

-0.68

afort

-0.67

case

-0.67

cember

-0.65

nutrition

-0.64

POSITIVE LOGITS

TW

1.05

ILD

1.04

SW

0.93

Ts

0.93

EEK

0.91

atcher

0.91

GW

0.89

roups

0.89

uild

0.89

KR

0.87

Activations Density 0.023%

references to power capacity measurements in gigawatts (GW)

mentions of specific geographical locations and entities

No Comments

No Known Activations