INDEX

Explanations

the name "Ross" with varying degrees of emphasis

oai_token-act-pair · gpt-3.5-turbo

mentions of a specific name or term related to "Gross."

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

 uninsured

-0.71

 impaired

-0.64

 loan

-0.64

 nerv

-0.63

 tint

-0.62

©¶æ

-0.62

 Peninsula

-0.62

 pains

-0.62

ãĥ£

-0.61

 marrow

-0.61

POSITIVE LOGITS

es

1.14

manship

1.09

ively

1.03

ional

1.02

ors

0.96

gow

0.95

=-=-

0.90

enger

0.87

words

0.85

engers

0.85

Activations Density 0.043%

the name "Ross" with varying degrees of emphasis

mentions of a specific name or term related to "Gross."

No Comments

No Known Activations