INDEX

Explanations

proper nouns, particularly names of companies, organizations, and individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to organizations or companies

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

 stabilization

-0.72

:]

-0.71

 cuff

-0.70

?]

-0.69

ifying

-0.67

!]

-0.66

 scrut

-0.66

 capacities

-0.65

heses

-0.64

:\

-0.64

POSITIVE LOGITS

aimon

0.73

onga

0.71

qt

0.67

 Mons

0.67

oji

0.65

ickr

0.65

ONDON

0.64

VERTIS

0.64

nown

0.63

Laun

0.62

Activations Density 0.510%

proper nouns, particularly names of companies, organizations, and individuals

proper nouns related to organizations or companies

No Comments

No Known Activations