INDEX

Explanations

action verbs related to negative activities or events

oai_token-act-pair · gpt-3.5-turbo

actions associated with systemic disruption or control

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 grouping

-0.69

nton

-0.69

ijk

-0.68

say

-0.61

peria

-0.61

eka

-0.60

zip

-0.59

Discussion

-0.59

 confir

-0.58

 fitting

-0.58

POSITIVE LOGITS

GGGGGGGG

0.73

lass

0.73

redients

0.72

 ducks

0.71

Edge

0.70

edIn

0.68

 Squid

0.64

 rods

0.63

edge

0.63

 pige

0.62

Activations Density 0.353%

action verbs related to negative activities or events

actions associated with systemic disruption or control

No Comments

No Known Activations