INDEX

Explanations

phrases related to reasons or explanations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating reasons or explanations related to various subjects

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

lez

-0.83

load

-0.79

lyn

-0.70

mint

-0.70

iot

-0.69

melon

-0.68

jad

-0.67

tumblr

-0.65

 TABLE

-0.64

sey

-0.64

POSITIVE LOGITS

 neither

0.81

nor

0.80

 actionGroup

0.78

 unsub

0.75

 bureaucratic

0.75

 confidentiality

0.74

 pesky

0.74

 airspace

0.72

 insufficient

0.72

 nond

0.71

Activations Density 0.163%

phrases related to reasons or explanations

phrases indicating reasons or explanations related to various subjects

No Comments

No Known Activations