INDEX

Explanations

phrases indicating probability or likelihood

oai_token-act-pair · gpt-3.5-turbo

phrases related to probability and chances of outcomes

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

 ILCS

-0.87

minus

-0.86

geist

-0.84

raq

-0.77

heses

-0.76

abytes

-0.75

 istg

-0.74

 Extras

-0.74

olitical

-0.73

Sport

-0.73

POSITIVE LOGITS

 obtaining

1.22

 encountering

1.17

 getting

1.12

 acquiring

1.11

 succeeding

1.09

 completing

1.07

 achieving

1.04

 surviving

1.04

 winning

1.03

 becoming

1.02

Activations Density 0.101%

phrases indicating probability or likelihood

phrases related to probability and chances of outcomes

No Comments

No Known Activations