INDEX

Explanations

references to Franklin Roosevelt

oai_token-act-pair · gpt-3.5-turbo

references to Franklin D. Roosevelt

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

akening

-0.86

rons

-0.81

ramid

-0.80

-0.79

oided

-0.79

ynt

-0.78

uana

-0.77

oslov

-0.75

untu

-0.74

los

-0.73

POSITIVE LOGITS

 Roosevelt

0.97

 Franklin

0.92

hurst

0.87

Fo

0.84

 Parkway

0.73

 Township

0.71

ãĥ¼ãĥĨ

0.70

 Contra

0.68

 Lakes

0.68

olini

0.68

Activations Density 0.012%

references to Franklin Roosevelt

references to Franklin D. Roosevelt

No Comments

No Known Activations