INDEX

Explanations

phrases related to someone's activities or achievements

oai_token-act-pair · gpt-3.5-turbo

the pronoun "He" in a variety of contexts

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

 Mandatory

-0.62

 guiActiveUnfocused

-0.62

INGTON

-0.61

earch

-0.61

SpaceEngineers

-0.58

âĻ¦

-0.57

Ø³

-0.56

Nar

-0.55

âĺħâĺħ

-0.54

packages

-0.54

POSITIVE LOGITS

resy

1.21

'll

1.13

'd

1.04

pherd

0.98

zbollah

0.96

reditary

0.91

ppard

0.90

've

0.83

fty

0.82

miah

0.81

Activations Density 0.238%

phrases related to someone's activities or achievements

the pronoun "He" in a variety of contexts

No Comments

No Known Activations