INDEX

Explanations

phrases related to offering assistance or help

oai_token-act-pair · gpt-3.5-turbo

actionable items or instructions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Sop

-0.79

 Info

-0.74

 Gadget

-0.73

Bot

-0.73

bush

-0.71

 weap

-0.70

 Tanzania

-0.69

spo

-0.68

 STATS

-0.67

 Sapp

-0.67

POSITIVE LOGITS

ul

0.99

ule

0.99

urion

0.95

NetMessage

0.92

uran

0.91

UL

0.91

catentry

0.91

ou

0.90

uld

0.87

actionDate

0.87

Activations Density 0.335%

phrases related to offering assistance or help

actionable items or instructions

No Comments

No Known Activations