INDEX

Explanations

phrases related to a specific name ("Ori")

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "ori" and its variants within various contexts

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

ted

-0.91

esville

-0.87

lier

-0.78

ting

-0.73

scl

-0.70

mom

-0.70

lycer

-0.70

dream

-0.69

TON

-0.69

driving

-0.67

POSITIVE LOGITS

Äĩ

1.06

arty

1.05

hei

0.96

Äį

0.89

pport

0.86

opsis

0.84

oti

0.83

wu

0.82

onto

0.79

ya

0.79

Activations Density 0.018%

phrases related to a specific name ("Ori")

occurrences of the term "ori" and its variants within various contexts

No Comments

No Known Activations