INDEX

Explanations

phrases related to events or actions involving multiple people or entities

oai_token-act-pair · gpt-3.5-turbo

phrases related to events involving groups, actions, and significant occurrences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 accordingly

-0.99

%.

-0.86

ãĢĤ

-0.83

 anyways

-0.83

*.

-0.81

 whereas

-0.79

 anyway

-0.77

+.

-0.77

 wherever

-0.75

etc

-0.74

POSITIVE LOGITS

pired

0.79

pires

0.73

 Collider

0.66

fired

0.63

ikh

0.61

tains

0.60

outed

0.58

 Loaded

0.57

ifies

0.57

named

0.57

Activations Density 0.503%

phrases related to events or actions involving multiple people or entities

phrases related to events involving groups, actions, and significant occurrences

No Comments

No Known Activations