INDEX

Explanations

verbs indicating observation, perception, or emphasis

oai_token-act-pair · gpt-3.5-turbo

statements that assert certainty or clarity about a particular topic

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Delivery

-0.76

 Translation

-0.68

itures

-0.68

urances

-0.67

enium

-0.66

 Payments

-0.65

llor

-0.65

 Inventory

-0.65

heid

-0.64

 ..............

-0.64

POSITIVE LOGITS

 bothering

1.09

 noteworthy

1.01

 noticeable

1.01

 worrisome

0.97

 interesting

0.96

 important

0.93

 troubling

0.92

 pertinent

0.92

 overlooked

0.91

 missing

0.90

Activations Density 0.236%

verbs indicating observation, perception, or emphasis

statements that assert certainty or clarity about a particular topic

No Comments

No Known Activations