INDEX

Explanations

names or references related to a person named "He."

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of characters or significant figures

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 elig

-0.77

lished

-0.75

 ultras

-0.64

 Ultr

-0.63

 horr

-0.63

 bluff

-0.61

ADRA

-0.60

 nightmares

-0.57

clusions

-0.57

 Rivals

-0.57

POSITIVE LOGITS

igans

0.93

felt

0.87

kil

0.73

idan

0.73

frames

0.71

kens

0.71

chenko

0.69

ides

0.69

bal

0.67

kamp

0.67

Activations Density 0.083%

names or references related to a person named "He."

proper nouns, particularly names of characters or significant figures

No Comments

No Known Activations