INDEX

Explanations

specific details or instances mentioned in a text

oai_token-act-pair · gpt-3.5-turbo

references to the term "specific" across 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

 Tycoon

-0.76

 Feldman

-0.75

tackle

-0.70

ublic

-0.69

ulton

-0.68

AUT

-0.67

adays

-0.67

NP

-0.67

 ashore

-0.66

ĺħ

-0.66

POSITIVE LOGITS

ities

1.12

 identifiable

0.96

 geographic

0.89

 geographical

0.87

 wording

0.87

ties

0.87

 kinds

0.86

iveness

0.85

 embodiments

0.85

iations

0.81

Activations Density 0.022%

specific details or instances mentioned in a text

references to the term "specific" across various contexts

No Comments

No Known Activations