INDEX

Explanations

pronouns referring to individuals

oai_token-act-pair · gpt-3.5-turbo

references to pronouns indicating personal relationships and interactions

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

 Pens

-0.66

ucc

-0.61

 Glob

-0.60

UTC

-0.60

Esk

-0.60

 Mulcair

-0.59

 Nikol

-0.58

limits

-0.58

 Jindal

-0.57

hide

-0.56

POSITIVE LOGITS

own

0.89

selves

0.80

oneliness

0.76

reditary

0.76

abba

0.74

self

0.72

pherd

0.71

essage

0.71

OWN

0.70

heter

0.69

Activations Density 0.493%

pronouns referring to individuals

references to pronouns indicating personal relationships and interactions

No Comments

No Known Activations