INDEX

Explanations

words related to validity or verification

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'v'

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

ĪĴ

-0.90

¿½

-0.77

¥µ

-0.76

Ĥª

-0.68

maid

-0.68

--------------------------------------------------------

-0.67

£ı

-0.65

dress

-0.65

anguage

-0.64

EStream

-0.64

POSITIVE LOGITS

apor

1.18

ascular

1.16

irus

1.13

olution

1.13

intage

1.10

isions

1.06

arsity

1.04

ampire

1.04

irgin

1.04

ortex

1.02

Activations Density 0.035%

words related to validity or verification

occurrences of the letter 'v'

No Comments

No Known Activations