INDEX

Explanations

phrases related to movements or actions

oai_token-act-pair · gpt-3.5-turbo

specific actions or events related to personal experiences and milestones

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

far

-0.66

 notwithstanding

-0.66

 aver

-0.64

inen

-0.62

¶ħ

-0.62

 Nanto

-0.61

 exacerbated

-0.59

mong

-0.58

among

-0.58

POSITIVE LOGITS

uberty

0.86

enium

0.67

robe

0.65

],"

0.65

ascus

0.62

 exams

0.62

pection

0.62

ript

0.62

jen

0.60

azeera

0.60

Activations Density 0.852%

phrases related to movements or actions

specific actions or events related to personal experiences and milestones

No Comments

No Known Activations