INDEX

Explanations

verbs relating to actions or activities

oai_token-act-pair · gpt-3.5-turbo

action verbs or experiences

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

-0.76

................................................................

-0.70

ItemTracker

-0.67

FEC

-0.66

MpServer

-0.64

 forgetting

-0.63

****************

-0.63

>>\

-0.61

ãĥ´ãĤ¡

-0.61

 bund

-0.61

POSITIVE LOGITS

escape

0.93

ably

0.82

ingly

0.81

dress

0.81

heed

0.75

 theirs

0.74

able

0.70

hyde

0.70

ables

0.67

uate

0.67

Activations Density 0.334%

verbs relating to actions or activities

action verbs or experiences

No Comments

No Known Activations