INDEX

Explanations

words related to monetary terms or financial transactions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "in."

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

CLASSIFIED

-0.77

tein

-0.65

=]

-0.63

âĢ¢âĢ¢

-0.62

 rede

-0.60

¿½

-0.60

 motives

-0.58

CHA

-0.57

uthor

-0.57

Haunted

-0.56

POSITIVE LOGITS

 handy

0.98

ordinate

0.94

between

0.93

offensive

0.91

sole

0.90

versely

0.89

order

0.86

escap

0.85

vert

0.83

bound

0.80

Activations Density 0.078%

words related to monetary terms or financial transactions

occurrences of the word "in."

No Comments

No Known Activations