INDEX

Explanations

words related to products

oai_token-act-pair · gpt-3.5-turbo

references to products and their related specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

CLASSIFIED

-0.79

 Blaz

-0.74

trop

-0.69

roe

-0.67

 Leone

-0.67

QUEST

-0.66

asons

-0.64

 Brother

-0.64

forth

-0.64

throp

-0.62

POSITIVE LOGITS

ivity

1.32

ively

1.26

 packaging

1.09

ivities

1.04

iveness

1.00

 marketed

0.92

store

0.85

umers

0.85

 placement

0.84

ivism

0.83

Activations Density 0.051%

words related to products

references to products and their related specifications

No Comments

No Known Activations