INDEX

Explanations

phrases related to providing and receiving feedback

oai_token-act-pair · gpt-3.5-turbo

references to feedback and its role in improvement processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

wikipedia

-0.74

chin

-0.73

neys

-0.71

amina

-0.68

annis

-0.67

asar

-0.67

adem

-0.66

ffe

-0.65

eni

-0.64

 Mississ

-0.64

POSITIVE LOGITS

 feedback

1.12

 loops

0.96

 Feedback

0.92

 testers

0.86

ible

0.81

 loop

0.81

 assurance

0.78

 sessions

0.70

storms

0.70

 sugg

0.69

Activations Density 0.046%

phrases related to providing and receiving feedback

references to feedback and its role in improvement processes

No Comments

No Known Activations