INDEX

Explanations

the name "Sab" or "Sabrina" with varying levels of emphasis

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Sab" and its variations in different 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

opathy

-0.71

opathic

-0.71

uminati

-0.70

jamin

-0.70

ä

-0.68

culosis

-0.67

ä¸ī

-0.67

WAYS

-0.66

guyen

-0.66

 expectancy

-0.66

POSITIVE LOGITS

rina

1.03

rill

0.93

ras

0.91

eways

0.90

ril

0.90

aram

0.89

dar

0.87

arat

0.85

ulation

0.84

lers

0.83

Activations Density 0.014%

the name "Sab" or "Sabrina" with varying levels of emphasis

occurrences of the name "Sab" and its variations in different contexts

No Comments

No Known Activations