INDEX

Explanations

phrases or sentences that indicate collaboration or working together

oai_token-act-pair · gpt-3.5-turbo

instances of the word "together" 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 @ 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

Zam

-0.70

ble

-0.67

 typo

-0.60

null

-0.59

 Hornets

-0.59

dark

-0.58

 Simmons

-0.57

holes

-0.57

tsy

-0.57

anty

-0.57

POSITIVE LOGITS

ãĤ¤ãĥĪ

0.81

éĹĺ

0.72

NESS

0.70

åĳ

0.69

assisted

0.66

å§«

0.65

 prest

0.65

èĪ

0.64

isphere

0.64

ICA

0.63

Activations Density 0.017%

phrases or sentences that indicate collaboration or working together

instances of the word "together" in various contexts

No Comments

No Known Activations