INDEX

Explanations

phrases or words indicating a significant or impactful change or effect

oai_token-act-pair · gpt-3.5-turbo

the term "significantly" used to describe changes or impacts in various contexts

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

tein

-0.91

ifully

-0.80

nar

-0.78

iens

-0.74

eer

-0.73

elf

-0.72

elsen

-0.71

aily

-0.71

icip

-0.69

icion

-0.68

POSITIVE LOGITS

 impacting

1.01

 impacted

0.89

 effected

0.87

 deterior

0.81

 degrade

0.81

0.79

 affecting

0.77

 advant

0.77

 disrupt

0.77

 benefited

0.77

Activations Density 0.025%

phrases or words indicating a significant or impactful change or effect

the term "significantly" used to describe changes or impacts in various contexts

No Comments

No Known Activations