INDEX

Explanations

words with the suffix '-ing'

oai_token-act-pair · gpt-3.5-turbo

gerund forms of verbs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Leilan

-0.90

itars

-0.74

luaj

-0.71

 dimin

-0.68

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.67

WA

-0.64

SIGN

-0.62

 stun

-0.61

 Shooter

-0.61

 disarm

-0.59

POSITIVE LOGITS

redients

1.13

tons

1.05

athering

0.88

atio

0.84

ding

0.83

roup

0.82

stad

0.77

rad

0.76

ering

0.75

ling

0.75

Activations Density 0.058%

words with the suffix '-ing'

gerund forms of verbs

No Comments

No Known Activations