INDEX

Explanations

proper nouns or specific names

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "spotted" and related variations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

VOL

-0.65

 Volume

-0.63

priv

-0.63

consumer

-0.63

Priv

-0.62

Consumer

-0.61

Revolution

-0.60

ablishment

-0.60

rest

-0.60

Imm

-0.60

POSITIVE LOGITS

 spotted

3.58

 spotting

1.75

 noticed

1.66

 photographed

1.65

 detected

1.63

 discovered

1.44

 overheard

1.43

 unearthed

1.40

 uncovered

1.35

 flagged

1.35

Activations Density 0.012%

proper nouns or specific names

occurrences of the word "spotted" and related variations

No Comments

No Known Activations