INDEX

Explanations

terms related to dynamic elements or changeful contexts

oai_token-act-pair · gpt-3.5-turbo

references to dynamic concepts or environments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ã±

-0.90

chen

-0.88

hemat

-0.87

chens

-0.86

creen

-0.84

ablishment

-0.81

aku

-0.80

acular

-0.79

OUGH

-0.78

jamin

-0.78

POSITIVE LOGITS

 destro

0.92

 dynamic

0.90

ynamic

0.86

 dynamically

0.83

 dynamics

0.82

 streng

0.82

duo

0.80

Duo

0.78

 upkeep

0.77

 enhancement

0.76

Activations Density 0.006%

terms related to dynamic elements or changeful contexts

references to dynamic concepts or environments

No Comments

No Known Activations