INDEX

Explanations

phrases or terms indicating restriction or limitation to a specific group or purpose

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate restrictions or limitations on use

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

ahime

-0.67

idon

-0.60

insula

-0.59

raught

-0.58

duino

-0.58

 Nurs

-0.57

illin

-0.57

 PLUS

-0.57

env

-0.56

anon

-0.55

POSITIVE LOGITS

 marginally

0.97

ices

0.94

ICES

0.84

 spor

0.80

 incidentally

0.77

 kidding

0.75

 insofar

0.74

 lasts

0.73

 scratches

0.72

onse

0.71

Activations Density 0.059%

phrases or terms indicating restriction or limitation to a specific group or purpose

phrases that indicate restrictions or limitations on use

No Comments

No Known Activations