INDEX

Explanations

the word "sent to" followed by a location or place

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "sent to" and related variations

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

 ramifications

-0.80

 implications

-0.70

natureconservancy

-0.65

ById

-0.64

 complexities

-0.64

 possibilities

-0.62

 dictates

-0.61

margin

-0.61

 minded

-0.61

 repercussions

-0.61

POSITIVE LOGITS

pload

1.09

 inspect

0.84

 relieve

0.82

 compensate

0.79

 replace

0.78

 Mandal

0.75

pless

0.75

ilet

0.75

 orbit

0.75

 Bangkok

0.74

Activations Density 0.226%

the word "sent to" followed by a location or place

instances of the phrase "sent to" and related variations

No Comments

No Known Activations