INDEX

Explanations

phrases related to peaceful intentions or activities

oai_token-act-pair · gpt-3.5-turbo

references to peacefulness in various contexts

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

ails

-0.80

è¦ļéĨĴ

-0.73

olls

-0.71

MAC

-0.69

odor

-0.67

 nause

-0.67

ulas

-0.66

mal

-0.66

ravings

-0.65

oglobin

-0.65

POSITIVE LOGITS

 peaceful

0.87

 minded

0.83

peace

0.78

ness

0.77

\\\\\\\\

0.76

edIn

0.76

keeping

0.74

haven

0.74

 disarm

0.73

edom

0.73

Activations Density 0.019%

phrases related to peaceful intentions or activities

references to peacefulness in various contexts

No Comments

No Known Activations