INDEX

Explanations

brand names and product names

oai_token-act-pair · gpt-3.5-turbo

brands and types of products or services

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

rawdownloadcloneembedreportprint

-0.84

REDACTED

-0.73

 Helpful

-0.66

VERTISEMENT

-0.66

cific

-0.64

lihood

-0.63

ĸļ

-0.62

nown

-0.61

abama

-0.59

 fixme

-0.59

POSITIVE LOGITS

ops

0.61

markets

0.59

 adapters

0.59

tech

0.59

ums

0.58

 warehouses

0.56

queue

0.56

 metic

0.55

 malls

0.55

 collaps

0.54

Activations Density 0.901%

brand names and product names

brands and types of products or services

No Comments

No Known Activations