INDEX

Explanations

phrases related to feedback and reactions, including their positive and negative aspects

oai_token-act-pair · gpt-3.5-turbo

terms related to negative feedback and its associated emotional reactions

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

ulhu

-0.77

ortium

-0.71

 refin

-0.70

igham

-0.70

prototype

-0.65

FN

-0.62

erey

-0.61

ADRA

-0.60

ritz

-0.60

olini

-0.60

POSITIVE LOGITS

 outweigh

0.78

 Syndrome

0.73

 outwe

0.72

 syndrome

0.65

 detract

0.64

 backdrop

0.63

thritis

0.62

agonist

0.61

 (âĪĴ

0.61

Ö¼

0.60

Activations Density 0.394%

phrases related to feedback and reactions, including their positive and negative aspects

terms related to negative feedback and its associated emotional reactions

No Comments

No Known Activations