INDEX

Explanations

the word "series"

oai_token-act-pair · gpt-3.5-turbo

references to sets of events or occurrences described as "series."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nai

-0.81

rella

-0.73

zan

-0.71

adel

-0.71

ossier

-0.69

laws

-0.69

angelo

-0.69

gary

-0.67

idge

-0.67

lesi

-0.65

POSITIVE LOGITS

ãĥĺ

0.71

encies

0.70

 finale

0.69

 Tactics

0.69

ĸļ

0.68

manship

0.68

 resistor

0.68

 Wire

0.67

 series

0.67

Timeout

0.65

Activations Density 0.025%

the word "series"

references to sets of events or occurrences described as "series."

No Comments

No Known Activations