INDEX

Explanations

proper nouns related to a specific named entity or person, possibly named "Ka"

oai_token-act-pair · gpt-3.5-turbo

references to a particular individual named "Ka."

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

IONS

-0.87

å§«

-0.83

Ö¼

-0.78

lessly

-0.78

cream

-0.78

å¤©

-0.77

ROR

-0.74

flies

-0.70

advertisement

-0.68

pheus

-0.67

POSITIVE LOGITS

iser

1.00

Ka

0.94

adian

0.94

umat

0.93

pling

0.92

isers

0.91

pler

0.91

iba

0.89

plan

0.88

Ka

0.87

Activations Density 0.016%

proper nouns related to a specific named entity or person, possibly named "Ka"

references to a particular individual named "Ka."

No Comments

No Known Activations