INDEX

Explanations

software-related terms and technical jargon

oai_token-act-pair · gpt-3.5-turbo

words related to technical specifications or components

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

 carbohyd

-0.71

 nursing

-0.67

 valve

-0.62

 bott

-0.61

ARI

-0.60

LET

-0.60

 fisher

-0.60

 forestry

-0.59

 judges

-0.58

 fathers

-0.58

POSITIVE LOGITS

hift

1.30

mith

1.25

ystem

1.21

hare

1.20

ource

1.20

ilver

1.20

peed

1.19

poons

1.14

cript

1.11

aurus

1.09

Activations Density 0.194%

software-related terms and technical jargon

words related to technical specifications or components

No Comments

No Known Activations