INDEX

Explanations

descriptions or characteristics of individuals

oai_token-act-pair · gpt-3.5-turbo

references to individuals characterized by their actions or roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

CRIP

-0.66

detail

-0.65

 Assessment

-0.63

Kin

-0.63

Plus

-0.62

 Checking

-0.62

 disadvant

-0.60

GREEN

-0.59

 Context

-0.58

ranging

-0.57

POSITIVE LOGITS

 loves

1.28

 specializes

1.28

 wears

1.26

 prefers

1.23

 hates

1.22

 enjoys

1.21

 likes

1.18

 refuses

1.18

 spends

1.13

 cares

1.13

Activations Density 0.112%

descriptions or characteristics of individuals

references to individuals characterized by their actions or roles

No Comments

No Known Activations