INDEX

Explanations

terms related to mistakes or misjudgments

oai_token-act-pair · gpt-3.5-turbo

terms related to misunderstandings or misrepresentations

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

 Fres

-0.68

minster

-0.67

ulhu

-0.67

Fra

-0.66

Lago

-0.64

winner

-0.61

MRI

-0.61

galitarian

-0.59

gel

-0.59

Ô

-0.58

POSITIVE LOGITS

vous

0.87

ceived

0.81

rued

0.75

gotten

0.73

fusc

0.71

 assumptions

0.70

WARE

0.69

ãĥ«

0.68

hoff

0.68

 misc

0.67

Activations Density 0.082%

terms related to mistakes or misjudgments

terms related to misunderstandings or misrepresentations

No Comments

No Known Activations