INDEX

Explanations

the word "object" and related forms

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "object" in 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

lla

-0.72

NPR

-0.71

ornia

-0.70

ricia

-0.70

artney

-0.67

elsius

-0.67

 millenn

-0.67

 Leone

-0.66

mingham

-0.66

corn

-0.64

POSITIVE LOGITS

ivity

1.15

ively

1.08

ivist

0.99

imus

0.94

ivism

0.93

ifying

0.92

ifies

0.92

ified

0.91

ion

0.90

ification

0.89

Activations Density 0.010%

the word "object" and related forms

occurrences of the word "object" in various contexts

No Comments

No Known Activations