INDEX

Explanations

references to ownership or possession of items

oai_token-act-pair · gpt-3.5-turbo

phrases related to ownership

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

antle

-0.82

romeda

-0.74

AFTA

-0.73

hesda

-0.73

bell

-0.69

anwhile

-0.69

————————————————

-0.68

anism

-0.67

reading

-0.67

————————

-0.66

POSITIVE LOGITS

 affairs

0.78

 accord

0.75

 hands

0.73

 fault

0.71

itiz

0.69

self

0.69

 branded

0.68

OWN

0.68

 proprietary

0.66

 version

0.66

Activations Density 0.037%

references to ownership or possession of items

phrases related to ownership

No Comments

No Known Activations