INDEX

Explanations

instances of the word "account" used in a context of providing information or a narrative

oai_token-act-pair · gpt-3.5-turbo

mentions of various "accounts" related to events or narratives

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

 Niet

-0.69

Aki

-0.63

NetMessage

-0.62

 cour

-0.62

JV

-0.61

 Unic

-0.61

 Sight

-0.61

GGGGGGGG

-0.60

beam

-0.59

Zip

-0.59

POSITIVE LOGITS

ancy

1.10

 rece

0.94

arily

0.91

abilities

0.89

ership

0.89

 accounts

0.89

 balances

0.88

Account

0.87

books

0.86

 payable

0.83

Activations Density 0.021%

instances of the word "account" used in a context of providing information or a narrative

mentions of various "accounts" related to events or narratives

No Comments

No Known Activations