INDEX

Explanations

mentions of previous instances or activities

oai_token-act-pair · gpt-3.5-turbo

references to prior events, studies, or efforts

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

ueller

-0.71

Pri

-0.69

 Genie

-0.67

wrapper

-0.63

$$$$

-0.63

,,,,

-0.59

 Horde

-0.58

liga

-0.58

Fans

-0.57

sung

-0.57

POSITIVE LOGITS

bernatorial

0.81

 administrations

0.79

arten

0.73

iations

0.72

 unsuccessfully

0.71

akedown

0.69

appa

0.65

ABE

0.65

 iterations

0.64

ean

0.64

Activations Density 0.229%

mentions of previous instances or activities

references to prior events, studies, or efforts

No Comments

No Known Activations