INDEX

Explanations

exclamations emphasizing positivity

oai_token-act-pair · gpt-3.5-turbo

expressions of strong emotional reactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 principally

-0.77

 denomin

-0.67

 anecd

-0.67

 Plaint

-0.67

 respondents

-0.67

iator

-0.66

env

-0.65

enhagen

-0.65

oint

-0.65

 plaintiffs

-0.65

POSITIVE LOGITS

 fuckin

1.15

!!!!!

1.11

 fucking

1.10

 gonna

1.09

!!!!

1.07

 Godd

1.06

!!!!!!!!

1.05

oooooooo

1.04

!!!

1.01

!!

0.99

Activations Density 0.511%

exclamations emphasizing positivity

expressions of strong emotional reactions

No Comments

No Known Activations