INDEX

Explanations

phrases related to positive qualities, morale, or strength

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "spirit" 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Printed

-0.81

 WATCHED

-0.72

PsyNetMessage

-0.69

tics

-0.67

DERR

-0.66

pse

-0.62

 Parenthood

-0.61

theless

-0.61

 Kessler

-0.57

rien

-0.57

POSITIVE LOGITS

uality

1.10

ual

0.92

uous

0.85

uum

0.74

atics

0.72

lio

0.72

atorium

0.72

lia

0.71

matic

0.70

ually

0.70

Activations Density 0.020%

phrases related to positive qualities, morale, or strength

references to the concept of "spirit" in various contexts

No Comments

No Known Activations