INDEX

Explanations

words or phrases related to interaction or engaging with others

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "interact" and related forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ft

-0.79

prus

-0.76

statement

-0.72

ften

-0.70

nown

-0.68

collection

-0.67

ffer

-0.67

aft

-0.67

ppelin

-0.66

enthal

-0.65

POSITIVE LOGITS

ivity

1.10

uate

1.02

 interactions

0.98

 interacts

0.96

 interact

0.94

 interacted

0.94

ively

0.93

ivities

0.89

uates

0.89

 interaction

0.87

Activations Density 0.012%

words or phrases related to interaction or engaging with others

occurrences of the word "interact" and related forms

No Comments

No Known Activations