INDEX

Explanations

specific named locations or organizations

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to significant places, events, or political entities

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

 Democr

-0.60

dylib

-0.54

indal

-0.53

rompt

-0.52

REDACTED

-0.52

IPS

-0.52

anwhile

-0.52

igree

-0.51

 clot

-0.50

igun

-0.50

POSITIVE LOGITS

 âĢº

0.58

 seiz

0.53

mania

0.49

 Aval

0.48

 Practices

0.47

0.46

Matrix

0.46

 Beaut

0.46

 resumes

0.45

Activations Density 1.543%

specific named locations or organizations

proper nouns related to significant places, events, or political entities

No Comments

No Known Activations