INDEX

Explanations

information related to individuals' birth details

oai_token-act-pair · gpt-3.5-turbo

phrases related to birth

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

phis

-0.73

awaru

-0.73

 tiss

-0.72

ahime

-0.71

vernment

-0.68

uyomi

-0.68

uko

-0.66

dden

-0.66

eredith

-0.66

allows

-0.64

POSITIVE LOGITS

lings

0.84

 born

0.79

days

0.75

stellar

0.74

 Born

0.70

 anew

0.70

born

0.69

ãĤ´

0.68

Ð»

0.67

ãĥĥãĤ¯

0.67

Activations Density 0.016%

information related to individuals' birth details

phrases related to birth

No Comments

No Known Activations