INDEX

Explanations

phrases related to results or outcomes

oai_token-act-pair · gpt-3.5-turbo

assertive statements and their outcomes in a context of results or conclusions

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

mods

-0.77

RPG

-0.75

soDeliveryDate

-0.71

inventoryQuantity

-0.70

Introduced

-0.69

 punishable

-0.65

auld

-0.65

 creeps

-0.65

inks

-0.62

roots

-0.61

POSITIVE LOGITS

Jol

0.75

ilon

0.71

agically

0.64

 preliminary

0.61

 sequencing

0.61

sym

0.60

 incon

0.59

 Shine

0.58

 conclusive

0.58

amount

0.57

Activations Density 0.290%

phrases related to results or outcomes

assertive statements and their outcomes in a context of results or conclusions

No Comments

No Known Activations