INDEX

Explanations

details or elements that are being deliberately omitted or neglected in a narrative or discussion

oai_token-act-pair · gpt-3.5-turbo

critical omissions or neglected points in discussions

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

Rating

-0.82

gran

-0.70

interrupted

-0.67

brother

-0.64

spr

-0.64

cu

-0.62

nis

-0.61

fu

-0.61

imen

-0.61

oub

-0.60

POSITIVE LOGITS

 nuance

0.80

 facts

0.79

 nuances

0.78

 mention

0.75

 explicitly

0.74

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.74

 Luxem

0.73

 acknowledge

0.73

 nuanced

0.72

 reality

0.70

Activations Density 0.354%

details or elements that are being deliberately omitted or neglected in a narrative or discussion

critical omissions or neglected points in discussions

No Comments

No Known Activations