INDEX

Explanations

references to information or data, such as dates, statistics, and records

oai_token-act-pair · gpt-3.5-turbo

references to data and statistics

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

":"/

-0.90

terms

-0.65

CHAT

-0.65

cav

-0.65

hid

-0.64

 quantity

-0.64

 absor

-0.63

autions

-0.63

cci

-0.63

 Parenthood

-0.62

POSITIVE LOGITS

liest

0.69

rys

0.67

wright

0.66

uper

0.66

 redes

0.66

keye

0.63

unes

0.62

ronic

0.62

Leg

0.62

 Normandy

0.61

Activations Density 0.000%

references to information or data, such as dates, statistics, and records

references to data and statistics

No Comments

No Known Activations