INDEX

Explanations

phrases related to similarity and mutual understanding

oai_token-act-pair · gpt-3.5-turbo

phrases that express preferences or opinions

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

MRI

-0.77

 cannabin

-0.73

rity

-0.72

ĸļ

-0.70

istries

-0.70

incial

-0.69

ornia

-0.68

osterone

-0.67

vation

-0.66

ãĥ´ãĤ¡

-0.65

POSITIVE LOGITS

ombies

0.72

 spoiled

0.70

catentry

0.61

zech

0.61

achus

0.60

ahime

0.60

 Ancients

0.59

minded

0.58

lins

0.57

illas

0.57

Activations Density 0.185%

phrases related to similarity and mutual understanding

phrases that express preferences or opinions

No Comments

No Known Activations