INDEX

Explanations

chemical compounds or related terms ending in "ene"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "ene" across 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

frames

-0.83

istically

-0.79

oons

-0.79

ties

-0.75

gerald

-0.74

istani

-0.73

ategory

-0.71

ricane

-0.70

RANT

-0.70

ernaut

-0.70

POSITIVE LOGITS

utics

1.18

uve

1.16

utical

1.10

alogy

0.97

 Wenger

0.94

quin

0.93

utic

0.90

volent

0.87

agles

0.87

issance

0.82

Activations Density 0.017%

chemical compounds or related terms ending in "ene"

occurrences of the substring "ene" across various contexts

No Comments

No Known Activations