INDEX

Explanations

phrases related to knowledge or insight

oai_token-act-pair · gpt-3.5-turbo

instances of the word "understanding" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

onies

-0.92

yss

-0.72

quer

-0.69

photos

-0.68

cow

-0.66

compan

-0.66

hire

-0.65

etheus

-0.65

inating

-0.64

 notor

-0.64

POSITIVE LOGITS

 Understanding

0.89

 understanding

0.83

ually

0.83

 comprehension

0.82

ably

0.82

 Sakuya

0.77

displayText

0.74

éĸ

0.72

FUL

0.72

 Understand

0.71

Activations Density 0.023%

phrases related to knowledge or insight

instances of the word "understanding" in various contexts

No Comments

No Known Activations