INDEX

Explanations

names of specific organizations or institutions

oai_token-act-pair · gpt-3.5-turbo

references to legal institutions and processes

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

iste

-0.98

istan

-0.87

 Nightmares

-0.86

itars

-0.72

psons

-0.70

rett

-0.70

 bunk

-0.67

bley

-0.67

istas

-0.66

 Doctrine

-0.66

POSITIVE LOGITS

CBI

0.82

GGGG

0.75

rating

0.73

FIR

0.70

Lens

0.69

 rooting

0.67

SHIP

0.67

rations

0.67

ONSORED

0.67

VD

0.66

Activations Density 0.038%

names of specific organizations or institutions

references to legal institutions and processes

No Comments

No Known Activations