INDEX

Explanations

specific names, particularly the name "Amanda"

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals, particularly those named Amanda

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

^^^^

-0.85

folk

-0.80

ÃŁ

-0.78

glers

-0.77

lda

-0.75

book

-0.72

^^

-0.68

ģĸ

-0.68

veland

-0.67

spr

-0.67

POSITIVE LOGITS

anth

0.83

ortion

0.75

atech

0.71

catentry

0.69

adic

0.69

Bio

0.69

atos

0.68

rophe

0.66

cia

0.66

algia

0.65

Activations Density 0.052%

specific names, particularly the name "Amanda"

references to specific individuals, particularly those named Amanda

No Comments

No Known Activations