INDEX

Explanations

terms related to enhancement or improvement

oai_token-act-pair · gpt-3.5-turbo

terms indicating improvement or enhancement 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

place

-0.76

cas

-0.74

metal

-0.69

ãĤ·ãĥ£

-0.66

gs

-0.65

ben

-0.65

sol

-0.65

gamer

-0.64

ndra

-0.64

POSITIVE LOGITS

ments

1.00

 enhancement

0.95

mble

0.90

 misunder

0.88

 enhancements

0.87

 confir

0.85

 enhances

0.84

 tremend

0.83

 ende

0.82

 enhancing

0.82

Activations Density 0.022%

terms related to enhancement or improvement

terms indicating improvement or enhancement in various contexts

No Comments

No Known Activations