INDEX

Explanations

names of programming languages and related technologies

oai_token-act-pair · gpt-3.5-turbo

keywords related to adverse effects and technical specifications

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

itionally

-0.79

ledge

-0.78

ting

-0.75

eworthy

-0.72

afort

-0.72

heet

-0.72

atform

-0.68

aminer

-0.68

oulos

-0.67

ainers

-0.67

POSITIVE LOGITS

ffect

0.81

UTH

0.73

vironment

0.71

Sov

0.68

lect

0.68

Xi

0.68

velop

0.67

 obser

0.67

anwhile

0.66

 Schwarz

0.66

Activations Density 0.042%

names of programming languages and related technologies

keywords related to adverse effects and technical specifications

No Comments

No Known Activations