INDEX

Explanations

terms related to customers or people attending an event

oai_token-act-pair · gpt-3.5-turbo

terms related to patrons and customers in various contexts

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

 Stalin

-0.67

 Marxism

-0.67

 Communist

-0.65

 Thing

-0.65

 Sabb

-0.64

 Workers

-0.64

 Pharma

-0.63

Reloaded

-0.62

 Genocide

-0.62

 Anthropology

-0.61

POSITIVE LOGITS

hip

1.10

aurus

0.94

 beware

0.94

opausal

0.91

erv

0.84

hips

0.81

uggest

0.81

 enjoy

0.80

ettings

0.79

 consume

0.79

Activations Density 0.204%

terms related to customers or people attending an event

terms related to patrons and customers in various contexts

No Comments

No Known Activations