INDEX

Explanations

places or locations described as 'nearby'

oai_token-act-pair · gpt-3.5-turbo

references to locations or places in close proximity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

icer

-0.85

inen

-0.81

udeb

-0.79

cher

-0.78

chin

-0.76

uid

-0.75

ocker

-0.73

rar

-0.73

UID

-0.73

UX

-0.73

POSITIVE LOGITS

 nearby

1.25

 neighbors

0.91

 neighboring

0.89

 adjacent

0.88

 neighbor

0.85

 vicinity

0.84

 neighbouring

0.82

 neighbours

0.77

 neighbour

0.77

 adjoining

0.77

Activations Density 0.007%

places or locations described as 'nearby'

references to locations or places in close proximity

No Comments

No Known Activations