INDEX

Explanations

phrases describing the ability or potential of something

oai_token-act-pair · gpt-3.5-turbo

expressions of capability or potential across 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

Moz

-0.65

 Yose

-0.65

cott

-0.61

 Meyer

-0.61

 Boots

-0.59

 Frag

-0.58

bard

-0.57

rule

-0.57

Soy

-0.57

cot

-0.57

POSITIVE LOGITS

bodied

0.96

nesses

0.96

icient

0.89

¶ħ

0.82

ive

0.76

 enough

0.74

lessly

0.74

itude

0.73

ively

0.72

 impairment

0.72

Activations Density 0.032%

phrases describing the ability or potential of something

expressions of capability or potential across various contexts

No Comments

No Known Activations