INDEX

Explanations

phrases related to enabling or facilitating actions for other individuals

oai_token-act-pair · gpt-3.5-turbo

phrases indicating capability or possibility associated with "they" or "he" actions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 irony

-0.82

 understatement

-0.74

ARK

-0.69

 setback

-0.61

 disbelief

-0.61

wikipedia

-0.60

 hypocrisy

-0.59

 contraction

-0.58

 Sagan

-0.57

++++++++++++++++

-0.56

POSITIVE LOGITS

sylv

0.90

ctic

0.73

pta

0.71

ptions

0.67

ividual

0.66

llular

0.66

burse

0.65

ilib

0.64

vre

0.64

 someday

0.64

Activations Density 0.119%

phrases related to enabling or facilitating actions for other individuals

phrases indicating capability or possibility associated with "they" or "he" actions

No Comments

No Known Activations