INDEX

Explanations

numeric sequences or combinations of letters and numbers

oai_token-act-pair · gpt-3.5-turbo

conjunctions and phrases indicating relationships or connections among items or concepts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

advertising

-0.77

 writ

-0.66

iosyncr

-0.64

 condem

-0.63

gem

-0.61

storage

-0.60

 expenses

-0.58

 fulfillment

-0.58

 afore

-0.57

waters

-0.57

POSITIVE LOGITS

CU

0.71

ENG

0.71

DK

0.71

 Gillespie

0.70

ALT

0.70

 Garfield

0.69

ÄŁ

0.68

xon

0.67

IDA

0.66

MAX

0.66

Activations Density 0.163%

numeric sequences or combinations of letters and numbers

conjunctions and phrases indicating relationships or connections among items or concepts

No Comments

No Known Activations