INDEX

Explanations

proper nouns or named entities

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to various organizations, locations, and notable individuals

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

 ..........

-0.77

iatrics

-0.71

lain

-0.71

staking

-0.70

kHz

-0.65

hement

-0.65

utils

-0.64

ĸļ

-0.62

steamapps

-0.61

 Tropical

-0.60

POSITIVE LOGITS

 citiz

0.75

sylv

0.70

 electorate

0.70

 populace

0.70

 workforce

0.67

istani

0.66

 contingent

0.65

 outing

0.65

 fanbase

0.64

 endors

0.64

Activations Density 0.441%

proper nouns or named entities

proper nouns related to various organizations, locations, and notable individuals

No Comments

No Known Activations