INDEX

Explanations

phrases related to lessons learned or sought after

oai_token-act-pair · gpt-3.5-turbo

terms related to lessons learned and outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

sit

-0.80

istries

-0.78

ĪĴ

-0.70

ateur

-0.70

alach

-0.70

abet

-0.70

hun

-0.70

atever

-0.67

riet

-0.67

utterstock

-0.66

POSITIVE LOGITS

ãĤ´

0.79

ãĤ´ãĥ³

0.75

 Measure

0.72

by

0.69

 herein

0.68

 uttered

0.67

 Procedure

0.67

esp

0.66

ãĥ¼ãĥĨ

0.65

 Mysteries

0.63

Activations Density 0.453%

phrases related to lessons learned or sought after

terms related to lessons learned and outcomes

No Comments

No Known Activations