INDEX

Explanations

proper nouns, specifically the surname "Lee"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Lee" 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 @ 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

rontal

-0.78

unct

-0.74

ALLY

-0.73

ionic

-0.69

ãĥ¼ãĥĨ

-0.69

ICAN

-0.68

icals

-0.66

ENTION

-0.64

security

-0.64

rians

-0.63

POSITIVE LOGITS

Anne

0.90

utenant

0.82

vel

0.82

warm

0.81

 Stras

0.80

 Chun

0.79

lee

0.78

gun

0.77

 Harvey

0.76

wei

0.75

Activations Density 0.025%

proper nouns, specifically the surname "Lee"

occurrences of the name "Lee" in various contexts

No Comments

No Known Activations