INDEX

Explanations

phrases indicating a sense of permanence or continuation in the present or future

oai_token-act-pair · gpt-3.5-turbo

frequently occurring adverbs and expressions indicating certainty or continuity

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

cos

-0.79

stad

-0.74

tics

-0.74

gdala

-0.71

lab

-0.67

quer

-0.66

Qual

-0.66

Sing

-0.66

regn

-0.65

role

-0.64

POSITIVE LOGITS

 gonna

0.90

been

0.73

 gotta

0.70

 Corner

0.70

got

0.68

 Plenty

0.67

 Chevy

0.67

 pretty

0.66

 Wrap

0.65

 Crazy

0.64

Activations Density 0.285%

phrases indicating a sense of permanence or continuation in the present or future

frequently occurring adverbs and expressions indicating certainty or continuity

No Comments

No Known Activations