INDEX

Explanations

descriptions of various incidents and events

oai_token-act-pair · gpt-3.5-turbo

references to specific cases or incidents involving notable events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

bern

-0.78

 isEnabled

-0.70

UME

-0.67

rients

-0.67

heit

-0.66

Lenin

-0.66

ionics

-0.65

tion

-0.64

Demand

-0.64

amines

-0.63

POSITIVE LOGITS

 recently

0.83

 unsuccessfully

0.75

 fiasco

0.74

 unsuccessful

0.72

 botched

0.71

 debacle

0.71

oÄŁ

0.70

 disastrous

0.70

 memorable

0.69

 swat

0.69

Activations Density 1.229%

descriptions of various incidents and events

references to specific cases or incidents involving notable events

No Comments

No Known Activations