INDEX

Explanations

questions and responses from a structured conversation

oai_token-act-pair · gpt-3.5-turbo

question formats and references to inquiries or prompts

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

 revel

-0.70

angering

-0.69

 outweigh

-0.69

 bloom

-0.68

 doomed

-0.68

utton

-0.68

aden

-0.67

 overshadow

-0.66

 stakes

-0.66

 culmin

-0.65

POSITIVE LOGITS

Hello

1.35

Hi

1.30

Hi

1.28

 Hello

1.25

reetings

1.14

Hey

1.08

Question

1.06

Dear

1.05

Hey

1.03

 Okay

1.02

Activations Density 0.138%

questions and responses from a structured conversation

question formats and references to inquiries or prompts

No Comments

No Known Activations