INDEX

Explanations

sentences containing the word "one" as the most emphasized term

oai_token-act-pair · gpt-3.5-turbo

repeated references to the word "one."

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

osponsors

-0.77

inders

-0.76

actionDate

-0.75

ategory

-0.67

inas

-0.63

anse

-0.63

ipers

-0.63

ãĤ¦ãĤ¹

-0.62

OAD

-0.62

ãĥ©ãĥ³

-0.61

POSITIVE LOGITS

 sided

0.81

 Hundred

0.79

 hundred

0.75

 Thousand

0.73

 thing

0.73

ones

0.72

esan

0.67

 iteration

0.66

 eyed

0.65

 Lone

0.65

Activations Density 0.032%

sentences containing the word "one" as the most emphasized term

repeated references to the word "one."

No Comments

No Known Activations