INDEX

Explanations

uncertainty and questioning phrases

oai_token-act-pair · gpt-3.5-turbo

questions reflecting uncertainty or confusion

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

å§«

-0.76

Discussion

-0.67

ests

-0.67

oir

-0.66

 sidx

-0.65

ribune

-0.65

ridor

-0.63

apses

-0.61

azar

-0.61

rites

-0.61

POSITIVE LOGITS

 anymore

0.79

nor

0.73

 spelling

0.71

 incent

0.68

 whereabouts

0.67

 acron

0.66

 Niet

0.64

 consolation

0.62

eve

0.62

 explan

0.62

Activations Density 0.117%

uncertainty and questioning phrases

questions reflecting uncertainty or confusion

No Comments

No Known Activations