INDEX

Explanations

descriptions of physical objects or actions, focusing on details or attributes

oai_token-act-pair · gpt-3.5-turbo

items and symbols that evoke nostalgia or references to the past

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 policies

-0.69

 settlements

-0.64

rosso

-0.62

 provinces

-0.60

 acad

-0.60

essions

-0.60

 principals

-0.59

 Universities

-0.59

ynes

-0.59

 Events

-0.58

POSITIVE LOGITS

 protr

0.72

xtap

0.67

washer

0.65

eus

0.64

 Razer

0.62

 circulated

0.61

shaped

0.60

paste

0.60

 symbol

0.60

Rot

0.58

Activations Density 1.367%

descriptions of physical objects or actions, focusing on details or attributes

items and symbols that evoke nostalgia or references to the past

No Comments

No Known Activations