INDEX

Explanations

specific recommendations mentioned in text

oai_token-act-pair · gpt-3.5-turbo

phrases related to recommendations or suggestions

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

atan

-0.75

inside

-0.72

adelphia

-0.72

Sus

-0.68

ker

-0.67

 Sovere

-0.67

 Brus

-0.67

thia

-0.67

kered

-0.66

erie

-0.65

POSITIVE LOGITS

Recommend

1.02

 recommendations

0.96

 recommendation

0.88

 recommending

0.85

recomm

0.78

 recommends

0.77

 recommended

0.72

ablo

0.70

 favorably

0.69

 guidelines

0.69

Activations Density 0.020%

specific recommendations mentioned in text

phrases related to recommendations or suggestions

No Comments

No Known Activations