INDEX

Explanations

phrases indicating a group or selection of specific items within a larger set

oai_token-act-pair · gpt-3.5-turbo

references to groups of people or entities 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

 Reincarn

-0.72

ahn

-0.62

orate

-0.59

 Appalachian

-0.57

 Civil

-0.57

 sorts

-0.57

 Intercept

-0.57

Comb

-0.56

Module

-0.55

 Fall

-0.55

POSITIVE LOGITS

ubes

0.86

ngth

0.85

ovych

0.71

awei

0.71

 seiz

0.71

Downloadha

0.70

 Micha

0.69

 challeng

0.67

axy

0.65

 favourites

0.64

Activations Density 0.058%

phrases indicating a group or selection of specific items within a larger set

references to groups of people or entities in various contexts

No Comments

No Known Activations