INDEX

Explanations

numerical values related to quantities of a hundred or similar words

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "hundred" in various numerical contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Nav

-0.72

Screen

-0.71

GROUND

-0.69

obal

-0.68

Land

-0.68

aker

-0.68

estine

-0.67

hra

-0.67

acey

-0.65

Chart

-0.64

POSITIVE LOGITS

 thousand

1.50

 eighty

1.11

 fifty

1.11

ths

1.07

 Thousand

1.05

 seventy

1.03

 dollars

1.02

 sixty

1.01

teenth

0.97

 hundred

0.96

Activations Density 0.047%

numerical values related to quantities of a hundred or similar words

occurrences of the word "hundred" in various numerical contexts

No Comments

No Known Activations