INDEX

Explanations

personal pronouns indicating ownership or involvement in a particular situation

oai_token-act-pair · gpt-3.5-turbo

pronouns referring to people or entities

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

hap

-0.67

pedia

-0.64

ipedia

-0.63

 Buddhism

-0.60

 endif

-0.59

Thank

-0.59

 Wake

-0.58

eah

-0.57

detail

-0.57

elta

-0.56

POSITIVE LOGITS

$,

0.60

saf

0.60

'll

0.58

 wont

0.58

tro

0.58

 streng

0.58

\",

0.58

 summoned

0.57

erie

0.56

eded

0.55

Activations Density 0.243%

personal pronouns indicating ownership or involvement in a particular situation

pronouns referring to people or entities

No Comments

No Known Activations