INDEX

Explanations

mentions of the word "Ent" or related to entities related to "Ent" in a variety of contexts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Ent" in 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

å£«

-0.78

 Archangel

-0.76

strap

-0.74

phrine

-0.72

 Spears

-0.70

manship

-0.67

BILITY

-0.65

 Jenner

-0.64

 sterling

-0.64

 Atmospheric

-0.62

POSITIVE LOGITS

ropy

1.21

itled

1.20

rance

1.12

renched

1.04

ipt

1.02

raction

1.01

itle

0.94

rants

0.94

ree

0.94

uri

0.93

Activations Density 0.010%

mentions of the word "Ent" or related to entities related to "Ent" in a variety of contexts

occurrences of the word "Ent" in various contexts

No Comments

No Known Activations