INDEX

Explanations

references to international organizations and diplomatic terms

oai_token-act-pair · gpt-3.5-turbo

references to the United Nations and its various agencies and councils

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

veland

-0.85

 Bucc

-0.80

osaurus

-0.73

omorph

-0.69

paio

-0.68

esville

-0.65

olla

-0.65

 combos

-0.65

eph

-0.63

oak

-0.63

POSITIVE LOGITS

assies

0.92

 intervene

0.88

 envoy

0.77

 recognised

0.75

 recognise

0.75

IFIED

0.73

 inspectors

0.71

 Luxembourg

0.71

recogn

0.71

iliation

0.70

Activations Density 0.178%

references to international organizations and diplomatic terms

references to the United Nations and its various agencies and councils

No Comments

No Known Activations